Duplicate from lunarring/latentblending

Co-authored-by: Johannes Stelzer <[email protected]>

.gitattributes

@@ -0,0 +1,34 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,13 @@
+---
+title: Latent Blending
+emoji: 🫠
+colorFrom: green
+colorTo: indigo
+sdk: gradio
+sdk_version: 3.19.1
+app_file: gradio_ui.py
+pinned: false
+duplicated_from: lunarring/latentblending
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

configs/v1-inference.yaml

@@ -0,0 +1,70 @@
+model:
+  base_learning_rate: 1.0e-04
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    linear_start: 0.00085
+    linear_end: 0.0120
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: "jpg"
+    cond_stage_key: "txt"
+    image_size: 64
+    channels: 4
+    cond_stage_trainable: false   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    use_ema: False
+
+    scheduler_config: # 10000 warmup steps
+      target: ldm.lr_scheduler.LambdaLinearScheduler
+      params:
+        warm_up_steps: [ 10000 ]
+        cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+        f_start: [ 1.e-6 ]
+        f_max: [ 1. ]
+        f_min: [ 1. ]
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        image_size: 32 # unused
+        in_channels: 4
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: True
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: true
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          num_res_blocks: 2
+          attn_resolutions: []
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenCLIPEmbedder

configs/v2-inference-v.yaml

@@ -0,0 +1,68 @@
+model:
+  base_learning_rate: 1.0e-4
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    parameterization: "v"
+    linear_start: 0.00085
+    linear_end: 0.0120
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: "jpg"
+    cond_stage_key: "txt"
+    image_size: 64
+    channels: 4
+    cond_stage_trainable: false
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    use_ema: False # we set this to false because this is an inference only config
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        use_checkpoint: True
+        use_fp16: True
+        image_size: 32 # unused
+        in_channels: 4
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_head_channels: 64 # need to fix for flash-attn
+        use_spatial_transformer: True
+        use_linear_in_transformer: True
+        transformer_depth: 1
+        context_dim: 1024
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          #attn_type: "vanilla-xformers"
+          double_z: true
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          num_res_blocks: 2
+          attn_resolutions: []
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenOpenCLIPEmbedder
+      params:
+        freeze: True
+        layer: "penultimate"

configs/v2-inference.yaml

@@ -0,0 +1,67 @@
+model:
+  base_learning_rate: 1.0e-4
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    linear_start: 0.00085
+    linear_end: 0.0120
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: "jpg"
+    cond_stage_key: "txt"
+    image_size: 64
+    channels: 4
+    cond_stage_trainable: false
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    use_ema: False # we set this to false because this is an inference only config
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        use_checkpoint: True
+        use_fp16: True
+        image_size: 32 # unused
+        in_channels: 4
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_head_channels: 64 # need to fix for flash-attn
+        use_spatial_transformer: True
+        use_linear_in_transformer: True
+        transformer_depth: 1
+        context_dim: 1024
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          #attn_type: "vanilla-xformers"
+          double_z: true
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          num_res_blocks: 2
+          attn_resolutions: []
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenOpenCLIPEmbedder
+      params:
+        freeze: True
+        layer: "penultimate"

configs/v2-inpainting-inference.yaml

@@ -0,0 +1,158 @@
+model:
+  base_learning_rate: 5.0e-05
+  target: ldm.models.diffusion.ddpm.LatentInpaintDiffusion
+  params:
+    linear_start: 0.00085
+    linear_end: 0.0120
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: "jpg"
+    cond_stage_key: "txt"
+    image_size: 64
+    channels: 4
+    cond_stage_trainable: false
+    conditioning_key: hybrid
+    scale_factor: 0.18215
+    monitor: val/loss_simple_ema
+    finetune_keys: null
+    use_ema: False
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        use_checkpoint: True
+        image_size: 32 # unused
+        in_channels: 9
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_head_channels: 64 # need to fix for flash-attn
+        use_spatial_transformer: True
+        use_linear_in_transformer: True
+        transformer_depth: 1
+        context_dim: 1024
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          #attn_type: "vanilla-xformers"
+          double_z: true
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+            - 1
+            - 2
+            - 4
+            - 4
+          num_res_blocks: 2
+          attn_resolutions: [ ]
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenOpenCLIPEmbedder
+      params:
+        freeze: True
+        layer: "penultimate"
+
+
+data:
+  target: ldm.data.laion.WebDataModuleFromConfig
+  params:
+    tar_base: null  # for concat as in LAION-A
+    p_unsafe_threshold: 0.1
+    filter_word_list: "data/filters.yaml"
+    max_pwatermark: 0.45
+    batch_size: 8
+    num_workers: 6
+    multinode: True
+    min_size: 512
+    train:
+      shards:
+        - "pipe:aws s3 cp s3://stability-aws/laion-a-native/part-0/{00000..18699}.tar -"
+        - "pipe:aws s3 cp s3://stability-aws/laion-a-native/part-1/{00000..18699}.tar -"
+        - "pipe:aws s3 cp s3://stability-aws/laion-a-native/part-2/{00000..18699}.tar -"
+        - "pipe:aws s3 cp s3://stability-aws/laion-a-native/part-3/{00000..18699}.tar -"
+        - "pipe:aws s3 cp s3://stability-aws/laion-a-native/part-4/{00000..18699}.tar -"  #{00000-94333}.tar"
+      shuffle: 10000
+      image_key: jpg
+      image_transforms:
+      - target: torchvision.transforms.Resize
+        params:
+          size: 512
+          interpolation: 3
+      - target: torchvision.transforms.RandomCrop
+        params:
+          size: 512
+      postprocess:
+        target: ldm.data.laion.AddMask
+        params:
+          mode: "512train-large"
+          p_drop: 0.25
+    # NOTE use enough shards to avoid empty validation loops in workers
+    validation:
+      shards:
+        - "pipe:aws s3 cp s3://deep-floyd-s3/datasets/laion_cleaned-part5/{93001..94333}.tar - "
+      shuffle: 0
+      image_key: jpg
+      image_transforms:
+      - target: torchvision.transforms.Resize
+        params:
+          size: 512
+          interpolation: 3
+      - target: torchvision.transforms.CenterCrop
+        params:
+          size: 512
+      postprocess:
+        target: ldm.data.laion.AddMask
+        params:
+          mode: "512train-large"
+          p_drop: 0.25
+
+lightning:
+  find_unused_parameters: True
+  modelcheckpoint:
+    params:
+      every_n_train_steps: 5000
+
+  callbacks:
+    metrics_over_trainsteps_checkpoint:
+      params:
+        every_n_train_steps: 10000
+
+    image_logger:
+      target: main.ImageLogger
+      params:
+        enable_autocast: False
+        disabled: False
+        batch_frequency: 1000
+        max_images: 4
+        increase_log_steps: False
+        log_first_step: False
+        log_images_kwargs:
+          use_ema_scope: False
+          inpaint: False
+          plot_progressive_rows: False
+          plot_diffusion_rows: False
+          N: 4
+          unconditional_guidance_scale: 5.0
+          unconditional_guidance_label: [""]
+          ddim_steps: 50  # todo check these out for depth2img,
+          ddim_eta: 0.0   # todo check these out for depth2img,
+
+  trainer:
+    benchmark: True
+    val_check_interval: 5000000
+    num_sanity_val_steps: 0
+    accumulate_grad_batches: 1

configs/v2-midas-inference.yaml

@@ -0,0 +1,74 @@
+model:
+  base_learning_rate: 5.0e-07
+  target: ldm.models.diffusion.ddpm.LatentDepth2ImageDiffusion
+  params:
+    linear_start: 0.00085
+    linear_end: 0.0120
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: "jpg"
+    cond_stage_key: "txt"
+    image_size: 64
+    channels: 4
+    cond_stage_trainable: false
+    conditioning_key: hybrid
+    scale_factor: 0.18215
+    monitor: val/loss_simple_ema
+    finetune_keys: null
+    use_ema: False
+
+    depth_stage_config:
+      target: ldm.modules.midas.api.MiDaSInference
+      params:
+        model_type: "dpt_hybrid"
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        use_checkpoint: True
+        image_size: 32 # unused
+        in_channels: 5
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_head_channels: 64 # need to fix for flash-attn
+        use_spatial_transformer: True
+        use_linear_in_transformer: True
+        transformer_depth: 1
+        context_dim: 1024
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          #attn_type: "vanilla-xformers"
+          double_z: true
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+            - 1
+            - 2
+            - 4
+            - 4
+          num_res_blocks: 2
+          attn_resolutions: [ ]
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenOpenCLIPEmbedder
+      params:
+        freeze: True
+        layer: "penultimate"
+
+

configs/x4-upscaling.yaml

@@ -0,0 +1,76 @@
+model:
+  base_learning_rate: 1.0e-04
+  target: ldm.models.diffusion.ddpm.LatentUpscaleDiffusion
+  params:
+    parameterization: "v"
+    low_scale_key: "lr"
+    linear_start: 0.0001
+    linear_end: 0.02
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: "jpg"
+    cond_stage_key: "txt"
+    image_size: 128
+    channels: 4
+    cond_stage_trainable: false
+    conditioning_key: "hybrid-adm"
+    monitor: val/loss_simple_ema
+    scale_factor: 0.08333
+    use_ema: False
+
+    low_scale_config:
+      target: ldm.modules.diffusionmodules.upscaling.ImageConcatWithNoiseAugmentation
+      params:
+        noise_schedule_config: # image space
+          linear_start: 0.0001
+          linear_end: 0.02
+        max_noise_level: 350
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        use_checkpoint: True
+        num_classes: 1000  # timesteps for noise conditioning (here constant, just need one)
+        image_size: 128
+        in_channels: 7
+        out_channels: 4
+        model_channels: 256
+        attention_resolutions: [ 2,4,8]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 2, 4]
+        disable_self_attentions: [True, True, True, False]
+        disable_middle_self_attn: False
+        num_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 1024
+        legacy: False
+        use_linear_in_transformer: True
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        ddconfig:
+          # attn_type: "vanilla-xformers" this model needs efficient attention to be feasible on HR data, also the decoder seems to break in half precision (UNet is fine though)
+          double_z: True
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult: [ 1,2,4 ]  # num_down = len(ch_mult)-1
+          num_res_blocks: 2
+          attn_resolutions: [ ]
+          dropout: 0.0
+
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenOpenCLIPEmbedder
+      params:
+        freeze: True
+        layer: "penultimate"
+

gradio_ui.py

@@ -0,0 +1,500 @@
+# Copyright 2022 Lunar Ring. All rights reserved.
+# Written by Johannes Stelzer, email [email protected] twitter @j_stelzer
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+import torch
+torch.backends.cudnn.benchmark = False
+torch.set_grad_enabled(False)
+import numpy as np
+import warnings
+warnings.filterwarnings('ignore')
+import warnings
+from tqdm.auto import tqdm
+from PIL import Image
+from movie_util import MovieSaver, concatenate_movies
+from latent_blending import LatentBlending
+from stable_diffusion_holder import StableDiffusionHolder
+import gradio as gr
+from dotenv import find_dotenv, load_dotenv
+import shutil
+import uuid
+from utils import get_time, add_frames_linear_interp
+from huggingface_hub import hf_hub_download
+
+
+class BlendingFrontend():
+    def __init__(
+            self,
+            sdh,
+            share=False):
+        r"""
+        Gradio Helper Class to collect UI data and start latent blending.
+        Args:
+            sdh:
+                StableDiffusionHolder
+            share: bool
+                Set true to get a shareable gradio link (e.g. for running a remote server)
+        """
+        self.share = share
+
+        # UI Defaults
+        self.num_inference_steps = 30
+        self.depth_strength = 0.25
+        self.seed1 = 420
+        self.seed2 = 420
+        self.prompt1 = ""
+        self.prompt2 = ""
+        self.negative_prompt = ""
+        self.fps = 30
+        self.duration_video = 8
+        self.t_compute_max_allowed = 10
+
+        self.lb = LatentBlending(sdh)
+        self.lb.sdh.num_inference_steps = self.num_inference_steps
+        self.init_parameters_from_lb()
+        self.init_save_dir()
+
+        # Vars
+        self.list_fp_imgs_current = []
+        self.recycle_img1 = False
+        self.recycle_img2 = False
+        self.list_all_segments = []
+        self.dp_session = ""
+        self.user_id = None
+
+    def init_parameters_from_lb(self):
+        r"""
+        Automatically init parameters from latentblending instance
+        """
+        self.height = self.lb.sdh.height
+        self.width = self.lb.sdh.width
+        self.guidance_scale = self.lb.guidance_scale
+        self.guidance_scale_mid_damper = self.lb.guidance_scale_mid_damper
+        self.mid_compression_scaler = self.lb.mid_compression_scaler
+        self.branch1_crossfeed_power = self.lb.branch1_crossfeed_power
+        self.branch1_crossfeed_range = self.lb.branch1_crossfeed_range
+        self.branch1_crossfeed_decay = self.lb.branch1_crossfeed_decay
+        self.parental_crossfeed_power = self.lb.parental_crossfeed_power
+        self.parental_crossfeed_range = self.lb.parental_crossfeed_range
+        self.parental_crossfeed_power_decay = self.lb.parental_crossfeed_power_decay
+
+    def init_save_dir(self):
+        r"""
+        Initializes the directory where stuff is being saved.
+        You can specify this directory in a ".env" file in your latentblending root, setting
+        DIR_OUT='/path/to/saving'
+        """
+        load_dotenv(find_dotenv(), verbose=False)
+        self.dp_out = os.getenv("DIR_OUT")
+        if self.dp_out is None:
+            self.dp_out = ""
+        self.dp_imgs = os.path.join(self.dp_out, "imgs")
+        os.makedirs(self.dp_imgs, exist_ok=True)
+        self.dp_movies = os.path.join(self.dp_out, "movies")
+        os.makedirs(self.dp_movies, exist_ok=True)
+        self.save_empty_image()
+
+    def save_empty_image(self):
+        r"""
+        Saves an empty/black dummy image.
+        """
+        self.fp_img_empty = os.path.join(self.dp_imgs, 'empty.jpg')
+        Image.fromarray(np.zeros((self.height, self.width, 3), dtype=np.uint8)).save(self.fp_img_empty, quality=5)
+
+    def randomize_seed1(self):
+        r"""
+        Randomizes the first seed
+        """
+        seed = np.random.randint(0, 10000000)
+        self.seed1 = int(seed)
+        print(f"randomize_seed1: new seed = {self.seed1}")
+        return seed
+
+    def randomize_seed2(self):
+        r"""
+        Randomizes the second seed
+        """
+        seed = np.random.randint(0, 10000000)
+        self.seed2 = int(seed)
+        print(f"randomize_seed2: new seed = {self.seed2}")
+        return seed
+
+    def setup_lb(self, list_ui_vals):
+        r"""
+        Sets all parameters from the UI. Since gradio does not support to pass dictionaries,
+        we have to instead pass keys (list_ui_keys, global) and values (list_ui_vals)
+        """
+        # Collect latent blending variables
+        self.lb.set_width(list_ui_vals[list_ui_keys.index('width')])
+        self.lb.set_height(list_ui_vals[list_ui_keys.index('height')])
+        self.lb.set_prompt1(list_ui_vals[list_ui_keys.index('prompt1')])
+        self.lb.set_prompt2(list_ui_vals[list_ui_keys.index('prompt2')])
+        self.lb.set_negative_prompt(list_ui_vals[list_ui_keys.index('negative_prompt')])
+        self.lb.guidance_scale = list_ui_vals[list_ui_keys.index('guidance_scale')]
+        self.lb.guidance_scale_mid_damper = list_ui_vals[list_ui_keys.index('guidance_scale_mid_damper')]
+        self.t_compute_max_allowed = list_ui_vals[list_ui_keys.index('duration_compute')]
+        self.lb.num_inference_steps = list_ui_vals[list_ui_keys.index('num_inference_steps')]
+        self.lb.sdh.num_inference_steps = list_ui_vals[list_ui_keys.index('num_inference_steps')]
+        self.duration_video = list_ui_vals[list_ui_keys.index('duration_video')]
+        self.lb.seed1 = list_ui_vals[list_ui_keys.index('seed1')]
+        self.lb.seed2 = list_ui_vals[list_ui_keys.index('seed2')]
+        self.lb.branch1_crossfeed_power = list_ui_vals[list_ui_keys.index('branch1_crossfeed_power')]
+        self.lb.branch1_crossfeed_range = list_ui_vals[list_ui_keys.index('branch1_crossfeed_range')]
+        self.lb.branch1_crossfeed_decay = list_ui_vals[list_ui_keys.index('branch1_crossfeed_decay')]
+        self.lb.parental_crossfeed_power = list_ui_vals[list_ui_keys.index('parental_crossfeed_power')]
+        self.lb.parental_crossfeed_range = list_ui_vals[list_ui_keys.index('parental_crossfeed_range')]
+        self.lb.parental_crossfeed_power_decay = list_ui_vals[list_ui_keys.index('parental_crossfeed_power_decay')]
+        self.num_inference_steps = list_ui_vals[list_ui_keys.index('num_inference_steps')]
+        self.depth_strength = list_ui_vals[list_ui_keys.index('depth_strength')]
+
+        if len(list_ui_vals[list_ui_keys.index('user_id')]) > 1:
+            self.user_id = list_ui_vals[list_ui_keys.index('user_id')]
+        else:
+            # generate new user id
+            self.user_id = uuid.uuid4().hex
+            print(f"made new user_id: {self.user_id} at {get_time('second')}")
+
+    def save_latents(self, fp_latents, list_latents):
+        r"""
+        Saves a latent trajectory on disk, in npy format.
+        """
+        list_latents_cpu = [l.cpu().numpy() for l in list_latents]
+        np.save(fp_latents, list_latents_cpu)
+
+    def load_latents(self, fp_latents):
+        r"""
+        Loads a latent trajectory from disk, converts to torch tensor.
+        """
+        list_latents_cpu = np.load(fp_latents)
+        list_latents = [torch.from_numpy(l).to(self.lb.device) for l in list_latents_cpu]
+        return list_latents
+
+    def compute_img1(self, *args):
+        r"""
+        Computes the first transition image and returns it for display.
+        Sets all other transition images and last image to empty (as they are obsolete with this operation)
+        """
+        list_ui_vals = args
+        self.setup_lb(list_ui_vals)
+        fp_img1 = os.path.join(self.dp_imgs, f"img1_{self.user_id}")
+        img1 = Image.fromarray(self.lb.compute_latents1(return_image=True))
+        img1.save(fp_img1 + ".jpg")
+        self.save_latents(fp_img1 + ".npy", self.lb.tree_latents[0])
+        self.recycle_img1 = True
+        self.recycle_img2 = False
+        return [fp_img1 + ".jpg", self.fp_img_empty, self.fp_img_empty, self.fp_img_empty, self.fp_img_empty, self.user_id]
+
+    def compute_img2(self, *args):
+        r"""
+        Computes the last transition image and returns it for display.
+        Sets all other transition images to empty (as they are obsolete with this operation)
+        """
+        if not os.path.isfile(os.path.join(self.dp_imgs, f"img1_{self.user_id}.jpg")):  # don't do anything
+            return [self.fp_img_empty, self.fp_img_empty, self.fp_img_empty, self.fp_img_empty, self.user_id]
+        list_ui_vals = args
+        self.setup_lb(list_ui_vals)
+
+        self.lb.tree_latents[0] = self.load_latents(os.path.join(self.dp_imgs, f"img1_{self.user_id}.npy"))
+        fp_img2 = os.path.join(self.dp_imgs, f"img2_{self.user_id}")
+        img2 = Image.fromarray(self.lb.compute_latents2(return_image=True))
+        img2.save(fp_img2 + '.jpg')
+        self.save_latents(fp_img2 + ".npy", self.lb.tree_latents[-1])
+        self.recycle_img2 = True
+        # fixme save seeds. change filenames?
+        return [self.fp_img_empty, self.fp_img_empty, self.fp_img_empty, fp_img2 + ".jpg", self.user_id]
+
+    def compute_transition(self, *args):
+        r"""
+        Computes transition images and movie.
+        """
+        list_ui_vals = args
+        self.setup_lb(list_ui_vals)
+        print("STARTING TRANSITION...")
+        fixed_seeds = [self.seed1, self.seed2]
+        # Inject loaded latents (other user interference)
+        self.lb.tree_latents[0] = self.load_latents(os.path.join(self.dp_imgs, f"img1_{self.user_id}.npy"))
+        self.lb.tree_latents[-1] = self.load_latents(os.path.join(self.dp_imgs, f"img2_{self.user_id}.npy"))
+        imgs_transition = self.lb.run_transition(
+            recycle_img1=self.recycle_img1,
+            recycle_img2=self.recycle_img2,
+            num_inference_steps=self.num_inference_steps,
+            depth_strength=self.depth_strength,
+            t_compute_max_allowed=self.t_compute_max_allowed,
+            fixed_seeds=fixed_seeds)
+        print(f"Latent Blending pass finished ({get_time('second')}). Resulted in {len(imgs_transition)} images")
+
+        # Subselect three preview images
+        idx_img_prev = np.round(np.linspace(0, len(imgs_transition) - 1, 5)[1:-1]).astype(np.int32)
+
+        list_imgs_preview = []
+        for j in idx_img_prev:
+            list_imgs_preview.append(Image.fromarray(imgs_transition[j]))
+
+        # Save the preview imgs as jpgs on disk so we are not sending umcompressed data around
+        current_timestamp = get_time('second')
+        self.list_fp_imgs_current = []
+        for i in range(len(list_imgs_preview)):
+            fp_img = os.path.join(self.dp_imgs, f"img_preview_{i}_{current_timestamp}.jpg")
+            list_imgs_preview[i].save(fp_img)
+            self.list_fp_imgs_current.append(fp_img)
+        # Insert cheap frames for the movie
+        imgs_transition_ext = add_frames_linear_interp(imgs_transition, self.duration_video, self.fps)
+
+        # Save as movie
+        self.fp_movie = self.get_fp_video_last()
+        if os.path.isfile(self.fp_movie):
+            os.remove(self.fp_movie)
+        ms = MovieSaver(self.fp_movie, fps=self.fps)
+        for img in tqdm(imgs_transition_ext):
+            ms.write_frame(img)
+        ms.finalize()
+        print("DONE SAVING MOVIE! SENDING BACK...")
+
+        # Assemble Output, updating the preview images and le movie
+        list_return = self.list_fp_imgs_current + [self.fp_movie]
+        return list_return
+
+    def stack_forward(self, prompt2, seed2):
+        r"""
+        Allows to generate multi-segment movies. Sets last image -> first image with all
+        relevant parameters.
+        """
+        # Save preview images, prompts and seeds into dictionary for stacking
+        if len(self.list_all_segments) == 0:
+            timestamp_session = get_time('second')
+            self.dp_session = os.path.join(self.dp_out, f"session_{timestamp_session}")
+            os.makedirs(self.dp_session)
+
+        idx_segment = len(self.list_all_segments)
+        dp_segment = os.path.join(self.dp_session, f"segment_{str(idx_segment).zfill(3)}")
+
+        self.list_all_segments.append(dp_segment)
+        self.lb.write_imgs_transition(dp_segment)
+
+        fp_movie_last = self.get_fp_video_last()
+        fp_movie_next = self.get_fp_video_next()
+
+        shutil.copyfile(fp_movie_last, fp_movie_next)
+
+        self.lb.tree_latents[0] = self.load_latents(os.path.join(self.dp_imgs, f"img1_{self.user_id}.npy"))
+        self.lb.tree_latents[-1] = self.load_latents(os.path.join(self.dp_imgs, f"img2_{self.user_id}.npy"))
+        self.lb.swap_forward()
+
+        shutil.copyfile(os.path.join(self.dp_imgs, f"img2_{self.user_id}.npy"), os.path.join(self.dp_imgs, f"img1_{self.user_id}.npy"))
+        fp_multi = self.multi_concat()
+        list_out = [fp_multi]
+
+        list_out.extend([os.path.join(self.dp_imgs, f"img2_{self.user_id}.jpg")])
+        list_out.extend([self.fp_img_empty] * 4)
+        list_out.append(gr.update(interactive=False, value=prompt2))
+        list_out.append(gr.update(interactive=False, value=seed2))
+        list_out.append("")
+        list_out.append(np.random.randint(0, 10000000))
+        print(f"stack_forward: fp_multi {fp_multi}")
+        return list_out
+
+    def multi_concat(self):
+        r"""
+        Concatentates all stacked segments into one long movie.
+        """
+        list_fp_movies = self.get_fp_video_all()
+        # Concatenate movies and save
+        fp_final = os.path.join(self.dp_session, f"concat_{self.user_id}.mp4")
+        concatenate_movies(fp_final, list_fp_movies)
+        return fp_final
+
+    def get_fp_video_all(self):
+        r"""
+        Collects all stacked movie segments.
+        """
+        list_all = os.listdir(self.dp_movies)
+        str_beg = f"movie_{self.user_id}_"
+        list_user = [l for l in list_all if str_beg in l]
+        list_user.sort()
+        list_user = [os.path.join(self.dp_movies, l) for l in list_user]
+        return list_user
+
+    def get_fp_video_next(self):
+        r"""
+        Gets the filepath of the next movie segment.
+        """
+        list_videos = self.get_fp_video_all()
+        if len(list_videos) == 0:
+            idx_next = 0
+        else:
+            idx_next = len(list_videos)
+        fp_video_next = os.path.join(self.dp_movies, f"movie_{self.user_id}_{str(idx_next).zfill(3)}.mp4")
+        return fp_video_next
+
+    def get_fp_video_last(self):
+        r"""
+        Gets the current video that was saved.
+        """
+        fp_video_last = os.path.join(self.dp_movies, f"last_{self.user_id}.mp4")
+        return fp_video_last
+
+
+if __name__ == "__main__":
+    fp_ckpt = hf_hub_download(repo_id="stabilityai/stable-diffusion-2-1-base", filename="v2-1_512-ema-pruned.ckpt")
+    # fp_ckpt = hf_hub_download(repo_id="stabilityai/stable-diffusion-2-1", filename="v2-1_768-ema-pruned.ckpt")
+    bf = BlendingFrontend(StableDiffusionHolder(fp_ckpt))
+    # self = BlendingFrontend(None)
+
+    with gr.Blocks() as demo:
+        gr.HTML("""<h1>Latent Blending</h1>
+<p>Create butter-smooth transitions between prompts, powered by stable diffusion</p>
+<p>For faster inference without waiting in queue, you may duplicate the space and upgrade to GPU in settings.
+<br/>
+<a href="https://huggingface.co/spaces/lunarring/latentblending?duplicate=true">
+<img style="margin-top: 0em; margin-bottom: 0em" src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>
+</p>""")
+
+        with gr.Row():
+            prompt1 = gr.Textbox(label="prompt 1")
+            prompt2 = gr.Textbox(label="prompt 2")
+
+        with gr.Row():
+            duration_compute = gr.Slider(10, 25, bf.t_compute_max_allowed, step=1, label='waiting time', interactive=True)
+            duration_video = gr.Slider(1, 100, bf.duration_video, step=0.1, label='video duration', interactive=True)
+            height = gr.Slider(256, 1024, bf.height, step=128, label='height', interactive=True)
+            width = gr.Slider(256, 1024, bf.width, step=128, label='width', interactive=True)
+
+        with gr.Accordion("Advanced Settings (click to expand)", open=False):
+
+            with gr.Accordion("Diffusion settings", open=True):
+                with gr.Row():
+                    num_inference_steps = gr.Slider(5, 100, bf.num_inference_steps, step=1, label='num_inference_steps', interactive=True)
+                    guidance_scale = gr.Slider(1, 25, bf.guidance_scale, step=0.1, label='guidance_scale', interactive=True)
+                    negative_prompt = gr.Textbox(label="negative prompt")
+
+            with gr.Accordion("Seed control: adjust seeds for first and last images", open=True):
+                with gr.Row():
+                    b_newseed1 = gr.Button("randomize seed 1", variant='secondary')
+                    seed1 = gr.Number(bf.seed1, label="seed 1", interactive=True)
+                    seed2 = gr.Number(bf.seed2, label="seed 2", interactive=True)
+                    b_newseed2 = gr.Button("randomize seed 2", variant='secondary')
+
+            with gr.Accordion("Last image crossfeeding.", open=True):
+                with gr.Row():
+                    branch1_crossfeed_power = gr.Slider(0.0, 1.0, bf.branch1_crossfeed_power, step=0.01, label='branch1 crossfeed power', interactive=True)
+                    branch1_crossfeed_range = gr.Slider(0.0, 1.0, bf.branch1_crossfeed_range, step=0.01, label='branch1 crossfeed range', interactive=True)
+                    branch1_crossfeed_decay = gr.Slider(0.0, 1.0, bf.branch1_crossfeed_decay, step=0.01, label='branch1 crossfeed decay', interactive=True)
+
+            with gr.Accordion("Transition settings", open=True):
+                with gr.Row():
+                    parental_crossfeed_power = gr.Slider(0.0, 1.0, bf.parental_crossfeed_power, step=0.01, label='parental crossfeed power', interactive=True)
+                    parental_crossfeed_range = gr.Slider(0.0, 1.0, bf.parental_crossfeed_range, step=0.01, label='parental crossfeed range', interactive=True)
+                    parental_crossfeed_power_decay = gr.Slider(0.0, 1.0, bf.parental_crossfeed_power_decay, step=0.01, label='parental crossfeed decay', interactive=True)
+                with gr.Row():
+                    depth_strength = gr.Slider(0.01, 0.99, bf.depth_strength, step=0.01, label='depth_strength', interactive=True)
+                    guidance_scale_mid_damper = gr.Slider(0.01, 2.0, bf.guidance_scale_mid_damper, step=0.01, label='guidance_scale_mid_damper', interactive=True)
+
+        with gr.Row():
+            b_compute1 = gr.Button('step1: compute first image', variant='primary')
+            b_compute2 = gr.Button('step2: compute last image', variant='primary')
+            b_compute_transition = gr.Button('step3: compute transition', variant='primary')
+
+        with gr.Row():
+            img1 = gr.Image(label="1/5")
+            img2 = gr.Image(label="2/5", show_progress=False)
+            img3 = gr.Image(label="3/5", show_progress=False)
+            img4 = gr.Image(label="4/5", show_progress=False)
+            img5 = gr.Image(label="5/5")
+
+        with gr.Row():
+            vid_single = gr.Video(label="current single trans")
+            vid_multi = gr.Video(label="concatented multi trans")
+
+        with gr.Row():
+            b_stackforward = gr.Button('append last movie segment (left) to multi movie (right)', variant='primary')
+
+        with gr.Row():
+            gr.Markdown(
+                """
+                # Parameters
+                ## Main
+                - waiting time: set your waiting time for the transition. high values = better quality
+                - video duration: seconds per segment
+                - height/width: in pixels
+
+                ## Diffusion settings
+                - num_inference_steps: number of diffusion steps
+                - guidance_scale: latent blending seems to prefer lower values here
+                - negative prompt: enter negative prompt here, applied for all images
+
+                ## Last image crossfeeding
+                - branch1_crossfeed_power: Controls the level of cross-feeding between the first and last image branch. For preserving structures.
+                - branch1_crossfeed_range: Sets the duration of active crossfeed during development. High values enforce strong structural similarity.
+                - branch1_crossfeed_decay: Sets decay for branch1_crossfeed_power. Lower values make the decay stronger across the range.
+
+                ## Transition settings
+                - parental_crossfeed_power: Similar to branch1_crossfeed_power, however applied for the images withinin the transition.
+                - parental_crossfeed_range: Similar to branch1_crossfeed_range, however applied for the images withinin the transition.
+                - parental_crossfeed_power_decay: Similar to branch1_crossfeed_decay, however applied for the images withinin the transition.
+                - depth_strength: Determines when the blending process will begin in terms of diffusion steps. Low values more inventive but can cause motion.
+                - guidance_scale_mid_damper: Decreases the guidance scale in the middle of a transition.
+                """)
+
+        with gr.Row():
+            user_id = gr.Textbox(label="user id", interactive=False)
+
+        # Collect all UI elemts in list to easily pass as inputs in gradio
+        dict_ui_elem = {}
+        dict_ui_elem["prompt1"] = prompt1
+        dict_ui_elem["negative_prompt"] = negative_prompt
+        dict_ui_elem["prompt2"] = prompt2
+
+        dict_ui_elem["duration_compute"] = duration_compute
+        dict_ui_elem["duration_video"] = duration_video
+        dict_ui_elem["height"] = height
+        dict_ui_elem["width"] = width
+
+        dict_ui_elem["depth_strength"] = depth_strength
+        dict_ui_elem["branch1_crossfeed_power"] = branch1_crossfeed_power
+        dict_ui_elem["branch1_crossfeed_range"] = branch1_crossfeed_range
+        dict_ui_elem["branch1_crossfeed_decay"] = branch1_crossfeed_decay
+
+        dict_ui_elem["num_inference_steps"] = num_inference_steps
+        dict_ui_elem["guidance_scale"] = guidance_scale
+        dict_ui_elem["guidance_scale_mid_damper"] = guidance_scale_mid_damper
+        dict_ui_elem["seed1"] = seed1
+        dict_ui_elem["seed2"] = seed2
+
+        dict_ui_elem["parental_crossfeed_range"] = parental_crossfeed_range
+        dict_ui_elem["parental_crossfeed_power"] = parental_crossfeed_power
+        dict_ui_elem["parental_crossfeed_power_decay"] = parental_crossfeed_power_decay
+        dict_ui_elem["user_id"] = user_id
+
+        # Convert to list, as gradio doesn't seem to accept dicts
+        list_ui_vals = []
+        list_ui_keys = []
+        for k in dict_ui_elem.keys():
+            list_ui_vals.append(dict_ui_elem[k])
+            list_ui_keys.append(k)
+        bf.list_ui_keys = list_ui_keys
+
+        b_newseed1.click(bf.randomize_seed1, outputs=seed1)
+        b_newseed2.click(bf.randomize_seed2, outputs=seed2)
+        b_compute1.click(bf.compute_img1, inputs=list_ui_vals, outputs=[img1, img2, img3, img4, img5, user_id])
+        b_compute2.click(bf.compute_img2, inputs=list_ui_vals, outputs=[img2, img3, img4, img5, user_id])
+        b_compute_transition.click(bf.compute_transition,
+                                   inputs=list_ui_vals,
+                                   outputs=[img2, img3, img4, vid_single])
+
+        b_stackforward.click(bf.stack_forward,
+                             inputs=[prompt2, seed2],
+                             outputs=[vid_multi, img1, img2, img3, img4, img5, prompt1, seed1, prompt2])
+
+    demo.launch(share=bf.share, inbrowser=True, inline=False)

latent_blending.py

@@ -0,0 +1,884 @@
+# Copyright 2022 Lunar Ring. All rights reserved.
+# Written by Johannes Stelzer, email [email protected] twitter @j_stelzer
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+import torch
+torch.backends.cudnn.benchmark = False
+torch.set_grad_enabled(False)
+import numpy as np
+import warnings
+warnings.filterwarnings('ignore')
+import time
+import warnings
+from tqdm.auto import tqdm
+from PIL import Image
+from movie_util import MovieSaver
+from typing import List, Optional
+from ldm.models.diffusion.ddpm import LatentUpscaleDiffusion, LatentInpaintDiffusion
+import lpips
+from utils import interpolate_spherical, interpolate_linear, add_frames_linear_interp, yml_load, yml_save
+
+
+class LatentBlending():
+    def __init__(
+            self,
+            sdh: None,
+            guidance_scale: float = 4,
+            guidance_scale_mid_damper: float = 0.5,
+            mid_compression_scaler: float = 1.2):
+        r"""
+        Initializes the latent blending class.
+        Args:
+            guidance_scale: float
+                Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
+                `guidance_scale` is defined as `w` of equation 2. of [Imagen
+                Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >
+                1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`,
+                usually at the expense of lower image quality.
+            guidance_scale_mid_damper: float = 0.5
+                Reduces the guidance scale towards the middle of the transition.
+                A value of 0.5 would decrease the guidance_scale towards the middle linearly by 0.5.
+            mid_compression_scaler: float = 2.0
+                Increases the sampling density in the middle (where most changes happen). Higher value
+                imply more values in the middle. However the inflection point can occur outside the middle,
+                thus high values can give rough transitions. Values around 2 should be fine.
+        """
+        assert guidance_scale_mid_damper > 0 \
+            and guidance_scale_mid_damper <= 1.0, \
+            f"guidance_scale_mid_damper neees to be in interval (0,1], you provided {guidance_scale_mid_damper}"
+
+        self.sdh = sdh
+        self.device = self.sdh.device
+        self.width = self.sdh.width
+        self.height = self.sdh.height
+        self.guidance_scale_mid_damper = guidance_scale_mid_damper
+        self.mid_compression_scaler = mid_compression_scaler
+        self.seed1 = 0
+        self.seed2 = 0
+
+        # Initialize vars
+        self.prompt1 = ""
+        self.prompt2 = ""
+        self.negative_prompt = ""
+
+        self.tree_latents = [None, None]
+        self.tree_fracts = None
+        self.idx_injection = []
+        self.tree_status = None
+        self.tree_final_imgs = []
+
+        self.list_nmb_branches_prev = []
+        self.list_injection_idx_prev = []
+        self.text_embedding1 = None
+        self.text_embedding2 = None
+        self.image1_lowres = None
+        self.image2_lowres = None
+        self.negative_prompt = None
+        self.num_inference_steps = self.sdh.num_inference_steps
+        self.noise_level_upscaling = 20
+        self.list_injection_idx = None
+        self.list_nmb_branches = None
+
+        # Mixing parameters
+        self.branch1_crossfeed_power = 0.1
+        self.branch1_crossfeed_range = 0.6
+        self.branch1_crossfeed_decay = 0.8
+
+        self.parental_crossfeed_power = 0.1
+        self.parental_crossfeed_range = 0.8
+        self.parental_crossfeed_power_decay = 0.8
+
+        self.set_guidance_scale(guidance_scale)
+        self.init_mode()
+        self.multi_transition_img_first = None
+        self.multi_transition_img_last = None
+        self.dt_per_diff = 0
+        self.spatial_mask = None
+        self.lpips = lpips.LPIPS(net='alex').cuda(self.device)
+
+    def init_mode(self):
+        r"""
+        Sets the operational mode. Currently supported are standard, inpainting and x4 upscaling.
+        """
+        if isinstance(self.sdh.model, LatentUpscaleDiffusion):
+            self.mode = 'upscale'
+        elif isinstance(self.sdh.model, LatentInpaintDiffusion):
+            self.sdh.image_source = None
+            self.sdh.mask_image = None
+            self.mode = 'inpaint'
+        else:
+            self.mode = 'standard'
+
+    def set_guidance_scale(self, guidance_scale):
+        r"""
+        sets the guidance scale.
+        """
+        self.guidance_scale_base = guidance_scale
+        self.guidance_scale = guidance_scale
+        self.sdh.guidance_scale = guidance_scale
+
+    def set_negative_prompt(self, negative_prompt):
+        r"""Set the negative prompt. Currenty only one negative prompt is supported
+        """
+        self.negative_prompt = negative_prompt
+        self.sdh.set_negative_prompt(negative_prompt)
+
+    def set_guidance_mid_dampening(self, fract_mixing):
+        r"""
+        Tunes the guidance scale down as a linear function of fract_mixing,
+        towards 0.5 the minimum will be reached.
+        """
+        mid_factor = 1 - np.abs(fract_mixing - 0.5) / 0.5
+        max_guidance_reduction = self.guidance_scale_base * (1 - self.guidance_scale_mid_damper) - 1
+        guidance_scale_effective = self.guidance_scale_base - max_guidance_reduction * mid_factor
+        self.guidance_scale = guidance_scale_effective
+        self.sdh.guidance_scale = guidance_scale_effective
+
+    def set_branch1_crossfeed(self, crossfeed_power, crossfeed_range, crossfeed_decay):
+        r"""
+        Sets the crossfeed parameters for the first branch to the last branch.
+        Args:
+            crossfeed_power: float [0,1]
+                Controls the level of cross-feeding between the first and last image branch.
+            crossfeed_range: float [0,1]
+                Sets the duration of active crossfeed during development.
+            crossfeed_decay: float [0,1]
+                Sets decay for branch1_crossfeed_power. Lower values make the decay stronger across the range.
+        """
+        self.branch1_crossfeed_power = np.clip(crossfeed_power, 0, 1)
+        self.branch1_crossfeed_range = np.clip(crossfeed_range, 0, 1)
+        self.branch1_crossfeed_decay = np.clip(crossfeed_decay, 0, 1)
+
+    def set_parental_crossfeed(self, crossfeed_power, crossfeed_range, crossfeed_decay):
+        r"""
+        Sets the crossfeed parameters for all transition images (within the first and last branch).
+        Args:
+            crossfeed_power: float [0,1]
+                Controls the level of cross-feeding from the parental branches
+            crossfeed_range: float [0,1]
+                Sets the duration of active crossfeed during development.
+            crossfeed_decay: float [0,1]
+                Sets decay for branch1_crossfeed_power. Lower values make the decay stronger across the range.
+        """
+        self.parental_crossfeed_power = np.clip(crossfeed_power, 0, 1)
+        self.parental_crossfeed_range = np.clip(crossfeed_range, 0, 1)
+        self.parental_crossfeed_power_decay = np.clip(crossfeed_decay, 0, 1)
+
+    def set_prompt1(self, prompt: str):
+        r"""
+        Sets the first prompt (for the first keyframe) including text embeddings.
+        Args:
+            prompt: str
+                ABC trending on artstation painted by Greg Rutkowski
+        """
+        prompt = prompt.replace("_", " ")
+        self.prompt1 = prompt
+        self.text_embedding1 = self.get_text_embeddings(self.prompt1)
+
+    def set_prompt2(self, prompt: str):
+        r"""
+        Sets the second prompt (for the second keyframe) including text embeddings.
+        Args:
+            prompt: str
+                XYZ trending on artstation painted by Greg Rutkowski
+        """
+        prompt = prompt.replace("_", " ")
+        self.prompt2 = prompt
+        self.text_embedding2 = self.get_text_embeddings(self.prompt2)
+
+    def set_image1(self, image: Image):
+        r"""
+        Sets the first image (keyframe), relevant for the upscaling model transitions.
+        Args:
+            image: Image
+        """
+        self.image1_lowres = image
+
+    def set_image2(self, image: Image):
+        r"""
+        Sets the second image (keyframe), relevant for the upscaling model transitions.
+        Args:
+            image: Image
+        """
+        self.image2_lowres = image
+
+    def run_transition(
+            self,
+            recycle_img1: Optional[bool] = False,
+            recycle_img2: Optional[bool] = False,
+            num_inference_steps: Optional[int] = 30,
+            depth_strength: Optional[float] = 0.3,
+            t_compute_max_allowed: Optional[float] = None,
+            nmb_max_branches: Optional[int] = None,
+            fixed_seeds: Optional[List[int]] = None):
+        r"""
+        Function for computing transitions.
+        Returns a list of transition images using spherical latent blending.
+        Args:
+            recycle_img1: Optional[bool]:
+                Don't recompute the latents for the first keyframe (purely prompt1). Saves compute.
+            recycle_img2: Optional[bool]:
+                Don't recompute the latents for the second keyframe (purely prompt2). Saves compute.
+            num_inference_steps:
+                Number of diffusion steps. Higher values will take more compute time.
+            depth_strength:
+                Determines how deep the first injection will happen.
+                Deeper injections will cause (unwanted) formation of new structures,
+                more shallow values will go into alpha-blendy land.
+            t_compute_max_allowed:
+                Either provide t_compute_max_allowed or nmb_max_branches.
+                The maximum time allowed for computation. Higher values give better results but take longer.
+            nmb_max_branches: int
+                Either provide t_compute_max_allowed or nmb_max_branches. The maximum number of branches to be computed. Higher values give better
+                results. Use this if you want to have controllable results independent
+                of your computer.
+            fixed_seeds: Optional[List[int)]:
+                You can supply two seeds that are used for the first and second keyframe (prompt1 and prompt2).
+                Otherwise random seeds will be taken.
+        """
+
+        # Sanity checks first
+        assert self.text_embedding1 is not None, 'Set the first text embedding with .set_prompt1(...) before'
+        assert self.text_embedding2 is not None, 'Set the second text embedding with .set_prompt2(...) before'
+
+        # Random seeds
+        if fixed_seeds is not None:
+            if fixed_seeds == 'randomize':
+                fixed_seeds = list(np.random.randint(0, 1000000, 2).astype(np.int32))
+            else:
+                assert len(fixed_seeds) == 2, "Supply a list with len = 2"
+
+            self.seed1 = fixed_seeds[0]
+            self.seed2 = fixed_seeds[1]
+
+        # Ensure correct num_inference_steps in holder
+        self.num_inference_steps = num_inference_steps
+        self.sdh.num_inference_steps = num_inference_steps
+
+        # Compute / Recycle first image
+        if not recycle_img1 or len(self.tree_latents[0]) != self.num_inference_steps:
+            list_latents1 = self.compute_latents1()
+        else:
+            list_latents1 = self.tree_latents[0]
+
+        # Compute / Recycle first image
+        if not recycle_img2 or len(self.tree_latents[-1]) != self.num_inference_steps:
+            list_latents2 = self.compute_latents2()
+        else:
+            list_latents2 = self.tree_latents[-1]
+
+        # Reset the tree, injecting the edge latents1/2 we just generated/recycled
+        self.tree_latents = [list_latents1, list_latents2]
+        self.tree_fracts = [0.0, 1.0]
+        self.tree_final_imgs = [self.sdh.latent2image((self.tree_latents[0][-1])), self.sdh.latent2image((self.tree_latents[-1][-1]))]
+        self.tree_idx_injection = [0, 0]
+
+        # Hard-fix. Apply spatial mask only for list_latents2 but not for transition. WIP...
+        self.spatial_mask = None
+
+        # Set up branching scheme (dependent on provided compute time)
+        list_idx_injection, list_nmb_stems = self.get_time_based_branching(depth_strength, t_compute_max_allowed, nmb_max_branches)
+
+        # Run iteratively, starting with the longest trajectory.
+        # Always inserting new branches where they are needed most according to image similarity
+        for s_idx in tqdm(range(len(list_idx_injection))):
+            nmb_stems = list_nmb_stems[s_idx]
+            idx_injection = list_idx_injection[s_idx]
+
+            for i in range(nmb_stems):
+                fract_mixing, b_parent1, b_parent2 = self.get_mixing_parameters(idx_injection)
+                self.set_guidance_mid_dampening(fract_mixing)
+                list_latents = self.compute_latents_mix(fract_mixing, b_parent1, b_parent2, idx_injection)
+                self.insert_into_tree(fract_mixing, idx_injection, list_latents)
+                # print(f"fract_mixing: {fract_mixing} idx_injection {idx_injection}")
+
+        return self.tree_final_imgs
+
+    def compute_latents1(self, return_image=False):
+        r"""
+        Runs a diffusion trajectory for the first image
+        Args:
+            return_image: bool
+                whether to return an image or the list of latents
+        """
+        print("starting compute_latents1")
+        list_conditionings = self.get_mixed_conditioning(0)
+        t0 = time.time()
+        latents_start = self.get_noise(self.seed1)
+        list_latents1 = self.run_diffusion(
+            list_conditionings,
+            latents_start=latents_start,
+            idx_start=0)
+        t1 = time.time()
+        self.dt_per_diff = (t1 - t0) / self.num_inference_steps
+        self.tree_latents[0] = list_latents1
+        if return_image:
+            return self.sdh.latent2image(list_latents1[-1])
+        else:
+            return list_latents1
+
+    def compute_latents2(self, return_image=False):
+        r"""
+        Runs a diffusion trajectory for the last image, which may be affected by the first image's trajectory.
+        Args:
+            return_image: bool
+                whether to return an image or the list of latents
+        """
+        print("starting compute_latents2")
+        list_conditionings = self.get_mixed_conditioning(1)
+        latents_start = self.get_noise(self.seed2)
+        # Influence from branch1
+        if self.branch1_crossfeed_power > 0.0:
+            # Set up the mixing_coeffs
+            idx_mixing_stop = int(round(self.num_inference_steps * self.branch1_crossfeed_range))
+            mixing_coeffs = list(np.linspace(self.branch1_crossfeed_power, self.branch1_crossfeed_power * self.branch1_crossfeed_decay, idx_mixing_stop))
+            mixing_coeffs.extend((self.num_inference_steps - idx_mixing_stop) * [0])
+            list_latents_mixing = self.tree_latents[0]
+            list_latents2 = self.run_diffusion(
+                list_conditionings,
+                latents_start=latents_start,
+                idx_start=0,
+                list_latents_mixing=list_latents_mixing,
+                mixing_coeffs=mixing_coeffs)
+        else:
+            list_latents2 = self.run_diffusion(list_conditionings, latents_start)
+        self.tree_latents[-1] = list_latents2
+
+        if return_image:
+            return self.sdh.latent2image(list_latents2[-1])
+        else:
+            return list_latents2
+
+    def compute_latents_mix(self, fract_mixing, b_parent1, b_parent2, idx_injection):
+        r"""
+        Runs a diffusion trajectory, using the latents from the respective parents
+        Args:
+            fract_mixing: float
+                the fraction along the transition axis [0, 1]
+            b_parent1: int
+                index of parent1 to be used
+            b_parent2: int
+                index of parent2 to be used
+            idx_injection: int
+                the index in terms of diffusion steps, where the next insertion will start.
+        """
+        list_conditionings = self.get_mixed_conditioning(fract_mixing)
+        fract_mixing_parental = (fract_mixing - self.tree_fracts[b_parent1]) / (self.tree_fracts[b_parent2] - self.tree_fracts[b_parent1])
+        # idx_reversed = self.num_inference_steps - idx_injection
+
+        list_latents_parental_mix = []
+        for i in range(self.num_inference_steps):
+            latents_p1 = self.tree_latents[b_parent1][i]
+            latents_p2 = self.tree_latents[b_parent2][i]
+            if latents_p1 is None or latents_p2 is None:
+                latents_parental = None
+            else:
+                latents_parental = interpolate_spherical(latents_p1, latents_p2, fract_mixing_parental)
+            list_latents_parental_mix.append(latents_parental)
+
+        idx_mixing_stop = int(round(self.num_inference_steps * self.parental_crossfeed_range))
+        mixing_coeffs = idx_injection * [self.parental_crossfeed_power]
+        nmb_mixing = idx_mixing_stop - idx_injection
+        if nmb_mixing > 0:
+            mixing_coeffs.extend(list(np.linspace(self.parental_crossfeed_power, self.parental_crossfeed_power * self.parental_crossfeed_power_decay, nmb_mixing)))
+        mixing_coeffs.extend((self.num_inference_steps - len(mixing_coeffs)) * [0])
+        latents_start = list_latents_parental_mix[idx_injection - 1]
+        list_latents = self.run_diffusion(
+            list_conditionings,
+            latents_start=latents_start,
+            idx_start=idx_injection,
+            list_latents_mixing=list_latents_parental_mix,
+            mixing_coeffs=mixing_coeffs)
+        return list_latents
+
+    def get_time_based_branching(self, depth_strength, t_compute_max_allowed=None, nmb_max_branches=None):
+        r"""
+        Sets up the branching scheme dependent on the time that is granted for compute.
+        The scheme uses an estimation derived from the first image's computation speed.
+        Either provide t_compute_max_allowed or nmb_max_branches
+        Args:
+            depth_strength:
+                Determines how deep the first injection will happen.
+                Deeper injections will cause (unwanted) formation of new structures,
+                more shallow values will go into alpha-blendy land.
+            t_compute_max_allowed: float
+                The maximum time allowed for computation. Higher values give better results
+                but take longer. Use this if you want to fix your waiting time for the results.
+            nmb_max_branches: int
+                The maximum number of branches to be computed. Higher values give better
+                results. Use this if you want to have controllable results independent
+                of your computer.
+        """
+        idx_injection_base = int(round(self.num_inference_steps * depth_strength))
+        list_idx_injection = np.arange(idx_injection_base, self.num_inference_steps - 1, 3)
+        list_nmb_stems = np.ones(len(list_idx_injection), dtype=np.int32)
+        t_compute = 0
+
+        if nmb_max_branches is None:
+            assert t_compute_max_allowed is not None, "Either specify t_compute_max_allowed or nmb_max_branches"
+            stop_criterion = "t_compute_max_allowed"
+        elif t_compute_max_allowed is None:
+            assert nmb_max_branches is not None, "Either specify t_compute_max_allowed or nmb_max_branches"
+            stop_criterion = "nmb_max_branches"
+            nmb_max_branches -= 2  # Discounting the outer frames
+        else:
+            raise ValueError("Either specify t_compute_max_allowed or nmb_max_branches")
+        stop_criterion_reached = False
+        is_first_iteration = True
+        while not stop_criterion_reached:
+            list_compute_steps = self.num_inference_steps - list_idx_injection
+            list_compute_steps *= list_nmb_stems
+            t_compute = np.sum(list_compute_steps) * self.dt_per_diff + 0.15 * np.sum(list_nmb_stems)
+            increase_done = False
+            for s_idx in range(len(list_nmb_stems) - 1):
+                if list_nmb_stems[s_idx + 1] / list_nmb_stems[s_idx] >= 2:
+                    list_nmb_stems[s_idx] += 1
+                    increase_done = True
+                    break
+            if not increase_done:
+                list_nmb_stems[-1] += 1
+
+            if stop_criterion == "t_compute_max_allowed" and t_compute > t_compute_max_allowed:
+                stop_criterion_reached = True
+            elif stop_criterion == "nmb_max_branches" and np.sum(list_nmb_stems) >= nmb_max_branches:
+                stop_criterion_reached = True
+                if is_first_iteration:
+                    # Need to undersample.
+                    list_idx_injection = np.linspace(list_idx_injection[0], list_idx_injection[-1], nmb_max_branches).astype(np.int32)
+                    list_nmb_stems = np.ones(len(list_idx_injection), dtype=np.int32)
+            else:
+                is_first_iteration = False
+
+            # print(f"t_compute {t_compute} list_nmb_stems {list_nmb_stems}")
+        return list_idx_injection, list_nmb_stems
+
+    def get_mixing_parameters(self, idx_injection):
+        r"""
+        Computes which parental latents should be mixed together to achieve a smooth blend.
+        As metric, we are using lpips image similarity. The insertion takes place
+        where the metric is maximal.
+        Args:
+            idx_injection: int
+                the index in terms of diffusion steps, where the next insertion will start.
+        """
+        # get_lpips_similarity
+        similarities = []
+        for i in range(len(self.tree_final_imgs) - 1):
+            similarities.append(self.get_lpips_similarity(self.tree_final_imgs[i], self.tree_final_imgs[i + 1]))
+        b_closest1 = np.argmax(similarities)
+        b_closest2 = b_closest1 + 1
+        fract_closest1 = self.tree_fracts[b_closest1]
+        fract_closest2 = self.tree_fracts[b_closest2]
+
+        # Ensure that the parents are indeed older!
+        b_parent1 = b_closest1
+        while True:
+            if self.tree_idx_injection[b_parent1] < idx_injection:
+                break
+            else:
+                b_parent1 -= 1
+        b_parent2 = b_closest2
+        while True:
+            if self.tree_idx_injection[b_parent2] < idx_injection:
+                break
+            else:
+                b_parent2 += 1
+        fract_mixing = (fract_closest1 + fract_closest2) / 2
+        return fract_mixing, b_parent1, b_parent2
+
+    def insert_into_tree(self, fract_mixing, idx_injection, list_latents):
+        r"""
+        Inserts all necessary parameters into the trajectory tree.
+        Args:
+            fract_mixing: float
+                the fraction along the transition axis [0, 1]
+            idx_injection: int
+                the index in terms of diffusion steps, where the next insertion will start.
+            list_latents: list
+                list of the latents to be inserted
+        """
+        b_parent1, b_parent2 = self.get_closest_idx(fract_mixing)
+        self.tree_latents.insert(b_parent1 + 1, list_latents)
+        self.tree_final_imgs.insert(b_parent1 + 1, self.sdh.latent2image(list_latents[-1]))
+        self.tree_fracts.insert(b_parent1 + 1, fract_mixing)
+        self.tree_idx_injection.insert(b_parent1 + 1, idx_injection)
+
+    def get_spatial_mask_template(self):
+        r"""
+        Experimental helper function to get a spatial mask template.
+        """
+        shape_latents = [self.sdh.C, self.sdh.height // self.sdh.f, self.sdh.width // self.sdh.f]
+        C, H, W = shape_latents
+        return np.ones((H, W))
+
+    def set_spatial_mask(self, img_mask):
+        r"""
+        Experimental helper function to set a spatial mask.
+        The mask forces latents to be overwritten.
+        Args:
+            img_mask:
+                mask image [0,1]. You can get a template using get_spatial_mask_template
+        """
+        shape_latents = [self.sdh.C, self.sdh.height // self.sdh.f, self.sdh.width // self.sdh.f]
+        C, H, W = shape_latents
+        img_mask = np.asarray(img_mask)
+        assert len(img_mask.shape) == 2, "Currently, only 2D images are supported as mask"
+        img_mask = np.clip(img_mask, 0, 1)
+        assert img_mask.shape[0] == H, f"Your mask needs to be of dimension {H} x {W}"
+        assert img_mask.shape[1] == W, f"Your mask needs to be of dimension {H} x {W}"
+        spatial_mask = torch.from_numpy(img_mask).to(device=self.device)
+        spatial_mask = torch.unsqueeze(spatial_mask, 0)
+        spatial_mask = spatial_mask.repeat((C, 1, 1))
+        spatial_mask = torch.unsqueeze(spatial_mask, 0)
+        self.spatial_mask = spatial_mask
+
+    def get_noise(self, seed):
+        r"""
+        Helper function to get noise given seed.
+        Args:
+            seed: int
+        """
+        generator = torch.Generator(device=self.sdh.device).manual_seed(int(seed))
+        if self.mode == 'standard':
+            shape_latents = [self.sdh.C, self.sdh.height // self.sdh.f, self.sdh.width // self.sdh.f]
+            C, H, W = shape_latents
+        elif self.mode == 'upscale':
+            w = self.image1_lowres.size[0]
+            h = self.image1_lowres.size[1]
+            shape_latents = [self.sdh.model.channels, h, w]
+            C, H, W = shape_latents
+        return torch.randn((1, C, H, W), generator=generator, device=self.sdh.device)
+
+    @torch.no_grad()
+    def run_diffusion(
+            self,
+            list_conditionings,
+            latents_start: torch.FloatTensor = None,
+            idx_start: int = 0,
+            list_latents_mixing=None,
+            mixing_coeffs=0.0,
+            return_image: Optional[bool] = False):
+        r"""
+        Wrapper function for diffusion runners.
+        Depending on the mode, the correct one will be executed.
+
+        Args:
+            list_conditionings: list
+                List of all conditionings for the diffusion model.
+            latents_start: torch.FloatTensor
+                Latents that are used for injection
+            idx_start: int
+                Index of the diffusion process start and where the latents_for_injection are injected
+            list_latents_mixing: torch.FloatTensor
+                List of latents (latent trajectories) that are used for mixing
+            mixing_coeffs: float or list
+                Coefficients, how strong each element of list_latents_mixing will be mixed in.
+            return_image: Optional[bool]
+                Optionally return image directly
+        """
+
+        # Ensure correct num_inference_steps in Holder
+        self.sdh.num_inference_steps = self.num_inference_steps
+        assert type(list_conditionings) is list, "list_conditionings need to be a list"
+
+        if self.mode == 'standard':
+            text_embeddings = list_conditionings[0]
+            return self.sdh.run_diffusion_standard(
+                text_embeddings=text_embeddings,
+                latents_start=latents_start,
+                idx_start=idx_start,
+                list_latents_mixing=list_latents_mixing,
+                mixing_coeffs=mixing_coeffs,
+                spatial_mask=self.spatial_mask,
+                return_image=return_image)
+
+        elif self.mode == 'upscale':
+            cond = list_conditionings[0]
+            uc_full = list_conditionings[1]
+            return self.sdh.run_diffusion_upscaling(
+                cond,
+                uc_full,
+                latents_start=latents_start,
+                idx_start=idx_start,
+                list_latents_mixing=list_latents_mixing,
+                mixing_coeffs=mixing_coeffs,
+                return_image=return_image)
+
+    def run_upscaling(
+            self,
+            dp_img: str,
+            depth_strength: float = 0.65,
+            num_inference_steps: int = 100,
+            nmb_max_branches_highres: int = 5,
+            nmb_max_branches_lowres: int = 6,
+            duration_single_segment=3,
+            fps=24,
+            fixed_seeds: Optional[List[int]] = None):
+        r"""
+        Runs upscaling with the x4 model. Requires that you run a transition before with a low-res model and save the results using write_imgs_transition.
+
+        Args:
+            dp_img: str
+                Path to the low-res transition path (as saved in write_imgs_transition)
+            depth_strength:
+                Determines how deep the first injection will happen.
+                Deeper injections will cause (unwanted) formation of new structures,
+                more shallow values will go into alpha-blendy land.
+            num_inference_steps:
+                Number of diffusion steps. Higher values will take more compute time.
+            nmb_max_branches_highres: int
+                Number of final branches of the upscaling transition pass. Note this is the number
+                of branches between each pair of low-res images.
+            nmb_max_branches_lowres: int
+                Number of input low-res images, subsampling all transition images written in the low-res pass.
+                Setting this number lower (e.g. 6) will decrease the compute time but not affect the results too much.
+            duration_single_segment: float
+                The duration of each high-res movie segment. You will have nmb_max_branches_lowres-1 segments in total.
+            fps: float
+                frames per second of movie
+            fixed_seeds: Optional[List[int)]:
+                You can supply two seeds that are used for the first and second keyframe (prompt1 and prompt2).
+                Otherwise random seeds will be taken.
+        """
+        fp_yml = os.path.join(dp_img, "lowres.yaml")
+        fp_movie = os.path.join(dp_img, "movie_highres.mp4")
+        ms = MovieSaver(fp_movie, fps=fps)
+        assert os.path.isfile(fp_yml), "lowres.yaml does not exist. did you forget run_upscaling_step1?"
+        dict_stuff = yml_load(fp_yml)
+
+        # load lowres images
+        nmb_images_lowres = dict_stuff['nmb_images']
+        prompt1 = dict_stuff['prompt1']
+        prompt2 = dict_stuff['prompt2']
+        idx_img_lowres = np.round(np.linspace(0, nmb_images_lowres - 1, nmb_max_branches_lowres)).astype(np.int32)
+        imgs_lowres = []
+        for i in idx_img_lowres:
+            fp_img_lowres = os.path.join(dp_img, f"lowres_img_{str(i).zfill(4)}.jpg")
+            assert os.path.isfile(fp_img_lowres), f"{fp_img_lowres} does not exist. did you forget run_upscaling_step1?"
+            imgs_lowres.append(Image.open(fp_img_lowres))
+
+        # set up upscaling
+        text_embeddingA = self.sdh.get_text_embedding(prompt1)
+        text_embeddingB = self.sdh.get_text_embedding(prompt2)
+        list_fract_mixing = np.linspace(0, 1, nmb_max_branches_lowres - 1)
+        for i in range(nmb_max_branches_lowres - 1):
+            print(f"Starting movie segment {i+1}/{nmb_max_branches_lowres-1}")
+            self.text_embedding1 = interpolate_linear(text_embeddingA, text_embeddingB, list_fract_mixing[i])
+            self.text_embedding2 = interpolate_linear(text_embeddingA, text_embeddingB, 1 - list_fract_mixing[i])
+            if i == 0:
+                recycle_img1 = False
+            else:
+                self.swap_forward()
+                recycle_img1 = True
+
+            self.set_image1(imgs_lowres[i])
+            self.set_image2(imgs_lowres[i + 1])
+
+            list_imgs = self.run_transition(
+                recycle_img1=recycle_img1,
+                recycle_img2=False,
+                num_inference_steps=num_inference_steps,
+                depth_strength=depth_strength,
+                nmb_max_branches=nmb_max_branches_highres)
+            list_imgs_interp = add_frames_linear_interp(list_imgs, fps, duration_single_segment)
+
+            # Save movie frame
+            for img in list_imgs_interp:
+                ms.write_frame(img)
+        ms.finalize()
+
+    @torch.no_grad()
+    def get_mixed_conditioning(self, fract_mixing):
+        if self.mode == 'standard':
+            text_embeddings_mix = interpolate_linear(self.text_embedding1, self.text_embedding2, fract_mixing)
+            list_conditionings = [text_embeddings_mix]
+        elif self.mode == 'inpaint':
+            text_embeddings_mix = interpolate_linear(self.text_embedding1, self.text_embedding2, fract_mixing)
+            list_conditionings = [text_embeddings_mix]
+        elif self.mode == 'upscale':
+            text_embeddings_mix = interpolate_linear(self.text_embedding1, self.text_embedding2, fract_mixing)
+            cond, uc_full = self.sdh.get_cond_upscaling(self.image1_lowres, text_embeddings_mix, self.noise_level_upscaling)
+            condB, uc_fullB = self.sdh.get_cond_upscaling(self.image2_lowres, text_embeddings_mix, self.noise_level_upscaling)
+            cond['c_concat'][0] = interpolate_spherical(cond['c_concat'][0], condB['c_concat'][0], fract_mixing)
+            uc_full['c_concat'][0] = interpolate_spherical(uc_full['c_concat'][0], uc_fullB['c_concat'][0], fract_mixing)
+            list_conditionings = [cond, uc_full]
+        else:
+            raise ValueError(f"mix_conditioning: unknown mode {self.mode}")
+        return list_conditionings
+
+    @torch.no_grad()
+    def get_text_embeddings(
+            self,
+            prompt: str):
+        r"""
+        Computes the text embeddings provided a string with a prompts.
+        Adapted from stable diffusion repo
+        Args:
+            prompt: str
+                ABC trending on artstation painted by Old Greg.
+        """
+        return self.sdh.get_text_embedding(prompt)
+
+    def write_imgs_transition(self, dp_img):
+        r"""
+        Writes the transition images into the folder dp_img.
+        Requires run_transition to be completed.
+        Args:
+            dp_img: str
+                Directory, into which the transition images, yaml file and latents are written.
+        """
+        imgs_transition = self.tree_final_imgs
+        os.makedirs(dp_img, exist_ok=True)
+        for i, img in enumerate(imgs_transition):
+            img_leaf = Image.fromarray(img)
+            img_leaf.save(os.path.join(dp_img, f"lowres_img_{str(i).zfill(4)}.jpg"))
+        fp_yml = os.path.join(dp_img, "lowres.yaml")
+        self.save_statedict(fp_yml)
+
+    def write_movie_transition(self, fp_movie, duration_transition, fps=30):
+        r"""
+        Writes the transition movie to fp_movie, using the given duration and fps..
+        The missing frames are linearly interpolated.
+        Args:
+            fp_movie: str
+                file pointer to the final movie.
+            duration_transition: float
+                duration of the movie in seonds
+            fps: int
+                fps of the movie
+        """
+
+        # Let's get more cheap frames via linear interpolation (duration_transition*fps frames)
+        imgs_transition_ext = add_frames_linear_interp(self.tree_final_imgs, duration_transition, fps)
+
+        # Save as MP4
+        if os.path.isfile(fp_movie):
+            os.remove(fp_movie)
+        ms = MovieSaver(fp_movie, fps=fps, shape_hw=[self.sdh.height, self.sdh.width])
+        for img in tqdm(imgs_transition_ext):
+            ms.write_frame(img)
+        ms.finalize()
+
+    def save_statedict(self, fp_yml):
+        # Dump everything relevant into yaml
+        imgs_transition = self.tree_final_imgs
+        state_dict = self.get_state_dict()
+        state_dict['nmb_images'] = len(imgs_transition)
+        yml_save(fp_yml, state_dict)
+
+    def get_state_dict(self):
+        state_dict = {}
+        grab_vars = ['prompt1', 'prompt2', 'seed1', 'seed2', 'height', 'width',
+                     'num_inference_steps', 'depth_strength', 'guidance_scale',
+                     'guidance_scale_mid_damper', 'mid_compression_scaler', 'negative_prompt',
+                     'branch1_crossfeed_power', 'branch1_crossfeed_range', 'branch1_crossfeed_decay'
+                     'parental_crossfeed_power', 'parental_crossfeed_range', 'parental_crossfeed_power_decay']
+        for v in grab_vars:
+            if hasattr(self, v):
+                if v == 'seed1' or v == 'seed2':
+                    state_dict[v] = int(getattr(self, v))
+                elif v == 'guidance_scale':
+                    state_dict[v] = float(getattr(self, v))
+
+                else:
+                    try:
+                        state_dict[v] = getattr(self, v)
+                    except Exception:
+                        pass
+        return state_dict
+
+    def randomize_seed(self):
+        r"""
+        Set a random seed for a fresh start.
+        """
+        seed = np.random.randint(999999999)
+        self.set_seed(seed)
+
+    def set_seed(self, seed: int):
+        r"""
+        Set a the seed for a fresh start.
+        """
+        self.seed = seed
+        self.sdh.seed = seed
+
+    def set_width(self, width):
+        r"""
+        Set the width of the resulting image.
+        """
+        assert np.mod(width, 64) == 0, "set_width: value needs to be divisible by 64"
+        self.width = width
+        self.sdh.width = width
+
+    def set_height(self, height):
+        r"""
+        Set the height of the resulting image.
+        """
+        assert np.mod(height, 64) == 0, "set_height: value needs to be divisible by 64"
+        self.height = height
+        self.sdh.height = height
+
+    def swap_forward(self):
+        r"""
+        Moves over keyframe two -> keyframe one. Useful for making a sequence of transitions
+        as in run_multi_transition()
+        """
+        # Move over all latents
+        self.tree_latents[0] = self.tree_latents[-1]
+        # Move over prompts and text embeddings
+        self.prompt1 = self.prompt2
+        self.text_embedding1 = self.text_embedding2
+        # Final cleanup for extra sanity
+        self.tree_final_imgs = []
+
+    def get_lpips_similarity(self, imgA, imgB):
+        r"""
+        Computes the image similarity between two images imgA and imgB.
+        Used to determine the optimal point of insertion to create smooth transitions.
+        High values indicate low similarity.
+        """
+        tensorA = torch.from_numpy(imgA).float().cuda(self.device)
+        tensorA = 2 * tensorA / 255.0 - 1
+        tensorA = tensorA.permute([2, 0, 1]).unsqueeze(0)
+        tensorB = torch.from_numpy(imgB).float().cuda(self.device)
+        tensorB = 2 * tensorB / 255.0 - 1
+        tensorB = tensorB.permute([2, 0, 1]).unsqueeze(0)
+        lploss = self.lpips(tensorA, tensorB)
+        lploss = float(lploss[0][0][0][0])
+        return lploss
+
+    # Auxiliary functions
+    def get_closest_idx(
+            self,
+            fract_mixing: float):
+        r"""
+        Helper function to retrieve the parents for any given mixing.
+        Example: fract_mixing = 0.4 and self.tree_fracts = [0, 0.3, 0.6, 1.0]
+        Will return the two closest values here, i.e. [1, 2]
+        """
+
+        pdist = fract_mixing - np.asarray(self.tree_fracts)
+        pdist_pos = pdist.copy()
+        pdist_pos[pdist_pos < 0] = np.inf
+        b_parent1 = np.argmin(pdist_pos)
+        pdist_neg = -pdist.copy()
+        pdist_neg[pdist_neg <= 0] = np.inf
+        b_parent2 = np.argmin(pdist_neg)
+
+        if b_parent1 > b_parent2:
+            tmp = b_parent2
+            b_parent2 = b_parent1
+            b_parent1 = tmp
+
+        return b_parent1, b_parent2

ldm/data/util.py

@@ -0,0 +1,24 @@
+import torch
+
+from ldm.modules.midas.api import load_midas_transform
+
+
+class AddMiDaS(object):
+    def __init__(self, model_type):
+        super().__init__()
+        self.transform = load_midas_transform(model_type)
+
+    def pt2np(self, x):
+        x = ((x + 1.0) * .5).detach().cpu().numpy()
+        return x
+
+    def np2pt(self, x):
+        x = torch.from_numpy(x) * 2 - 1.
+        return x
+
+    def __call__(self, sample):
+        # sample['jpg'] is tensor hwc in [-1, 1] at this point
+        x = self.pt2np(sample['jpg'])
+        x = self.transform({"image": x})["image"]
+        sample['midas_in'] = x
+        return sample

ldm/ldm

@@ -0,0 +1 @@
+ldm

ldm/models/autoencoder.py

@@ -0,0 +1,219 @@
+import torch
+import pytorch_lightning as pl
+import torch.nn.functional as F
+from contextlib import contextmanager
+
+from ldm.modules.diffusionmodules.model import Encoder, Decoder
+from ldm.modules.distributions.distributions import DiagonalGaussianDistribution
+
+from ldm.util import instantiate_from_config
+from ldm.modules.ema import LitEma
+
+
+class AutoencoderKL(pl.LightningModule):
+    def __init__(self,
+                 ddconfig,
+                 lossconfig,
+                 embed_dim,
+                 ckpt_path=None,
+                 ignore_keys=[],
+                 image_key="image",
+                 colorize_nlabels=None,
+                 monitor=None,
+                 ema_decay=None,
+                 learn_logvar=False
+                 ):
+        super().__init__()
+        self.learn_logvar = learn_logvar
+        self.image_key = image_key
+        self.encoder = Encoder(**ddconfig)
+        self.decoder = Decoder(**ddconfig)
+        self.loss = instantiate_from_config(lossconfig)
+        assert ddconfig["double_z"]
+        self.quant_conv = torch.nn.Conv2d(2*ddconfig["z_channels"], 2*embed_dim, 1)
+        self.post_quant_conv = torch.nn.Conv2d(embed_dim, ddconfig["z_channels"], 1)
+        self.embed_dim = embed_dim
+        if colorize_nlabels is not None:
+            assert type(colorize_nlabels)==int
+            self.register_buffer("colorize", torch.randn(3, colorize_nlabels, 1, 1))
+        if monitor is not None:
+            self.monitor = monitor
+
+        self.use_ema = ema_decay is not None
+        if self.use_ema:
+            self.ema_decay = ema_decay
+            assert 0. < ema_decay < 1.
+            self.model_ema = LitEma(self, decay=ema_decay)
+            print(f"Keeping EMAs of {len(list(self.model_ema.buffers()))}.")
+
+        if ckpt_path is not None:
+            self.init_from_ckpt(ckpt_path, ignore_keys=ignore_keys)
+
+    def init_from_ckpt(self, path, ignore_keys=list()):
+        sd = torch.load(path, map_location="cpu")["state_dict"]
+        keys = list(sd.keys())
+        for k in keys:
+            for ik in ignore_keys:
+                if k.startswith(ik):
+                    print("Deleting key {} from state_dict.".format(k))
+                    del sd[k]
+        self.load_state_dict(sd, strict=False)
+        print(f"Restored from {path}")
+
+    @contextmanager
+    def ema_scope(self, context=None):
+        if self.use_ema:
+            self.model_ema.store(self.parameters())
+            self.model_ema.copy_to(self)
+            if context is not None:
+                print(f"{context}: Switched to EMA weights")
+        try:
+            yield None
+        finally:
+            if self.use_ema:
+                self.model_ema.restore(self.parameters())
+                if context is not None:
+                    print(f"{context}: Restored training weights")
+
+    def on_train_batch_end(self, *args, **kwargs):
+        if self.use_ema:
+            self.model_ema(self)
+
+    def encode(self, x):
+        h = self.encoder(x)
+        moments = self.quant_conv(h)
+        posterior = DiagonalGaussianDistribution(moments)
+        return posterior
+
+    def decode(self, z):
+        z = self.post_quant_conv(z)
+        dec = self.decoder(z)
+        return dec
+
+    def forward(self, input, sample_posterior=True):
+        posterior = self.encode(input)
+        if sample_posterior:
+            z = posterior.sample()
+        else:
+            z = posterior.mode()
+        dec = self.decode(z)
+        return dec, posterior
+
+    def get_input(self, batch, k):
+        x = batch[k]
+        if len(x.shape) == 3:
+            x = x[..., None]
+        x = x.permute(0, 3, 1, 2).to(memory_format=torch.contiguous_format).float()
+        return x
+
+    def training_step(self, batch, batch_idx, optimizer_idx):
+        inputs = self.get_input(batch, self.image_key)
+        reconstructions, posterior = self(inputs)
+
+        if optimizer_idx == 0:
+            # train encoder+decoder+logvar
+            aeloss, log_dict_ae = self.loss(inputs, reconstructions, posterior, optimizer_idx, self.global_step,
+                                            last_layer=self.get_last_layer(), split="train")
+            self.log("aeloss", aeloss, prog_bar=True, logger=True, on_step=True, on_epoch=True)
+            self.log_dict(log_dict_ae, prog_bar=False, logger=True, on_step=True, on_epoch=False)
+            return aeloss
+
+        if optimizer_idx == 1:
+            # train the discriminator
+            discloss, log_dict_disc = self.loss(inputs, reconstructions, posterior, optimizer_idx, self.global_step,
+                                                last_layer=self.get_last_layer(), split="train")
+
+            self.log("discloss", discloss, prog_bar=True, logger=True, on_step=True, on_epoch=True)
+            self.log_dict(log_dict_disc, prog_bar=False, logger=True, on_step=True, on_epoch=False)
+            return discloss
+
+    def validation_step(self, batch, batch_idx):
+        log_dict = self._validation_step(batch, batch_idx)
+        with self.ema_scope():
+            log_dict_ema = self._validation_step(batch, batch_idx, postfix="_ema")
+        return log_dict
+
+    def _validation_step(self, batch, batch_idx, postfix=""):
+        inputs = self.get_input(batch, self.image_key)
+        reconstructions, posterior = self(inputs)
+        aeloss, log_dict_ae = self.loss(inputs, reconstructions, posterior, 0, self.global_step,
+                                        last_layer=self.get_last_layer(), split="val"+postfix)
+
+        discloss, log_dict_disc = self.loss(inputs, reconstructions, posterior, 1, self.global_step,
+                                            last_layer=self.get_last_layer(), split="val"+postfix)
+
+        self.log(f"val{postfix}/rec_loss", log_dict_ae[f"val{postfix}/rec_loss"])
+        self.log_dict(log_dict_ae)
+        self.log_dict(log_dict_disc)
+        return self.log_dict
+
+    def configure_optimizers(self):
+        lr = self.learning_rate
+        ae_params_list = list(self.encoder.parameters()) + list(self.decoder.parameters()) + list(
+            self.quant_conv.parameters()) + list(self.post_quant_conv.parameters())
+        if self.learn_logvar:
+            print(f"{self.__class__.__name__}: Learning logvar")
+            ae_params_list.append(self.loss.logvar)
+        opt_ae = torch.optim.Adam(ae_params_list,
+                                  lr=lr, betas=(0.5, 0.9))
+        opt_disc = torch.optim.Adam(self.loss.discriminator.parameters(),
+                                    lr=lr, betas=(0.5, 0.9))
+        return [opt_ae, opt_disc], []
+
+    def get_last_layer(self):
+        return self.decoder.conv_out.weight
+
+    @torch.no_grad()
+    def log_images(self, batch, only_inputs=False, log_ema=False, **kwargs):
+        log = dict()
+        x = self.get_input(batch, self.image_key)
+        x = x.to(self.device)
+        if not only_inputs:
+            xrec, posterior = self(x)
+            if x.shape[1] > 3:
+                # colorize with random projection
+                assert xrec.shape[1] > 3
+                x = self.to_rgb(x)
+                xrec = self.to_rgb(xrec)
+            log["samples"] = self.decode(torch.randn_like(posterior.sample()))
+            log["reconstructions"] = xrec
+            if log_ema or self.use_ema:
+                with self.ema_scope():
+                    xrec_ema, posterior_ema = self(x)
+                    if x.shape[1] > 3:
+                        # colorize with random projection
+                        assert xrec_ema.shape[1] > 3
+                        xrec_ema = self.to_rgb(xrec_ema)
+                    log["samples_ema"] = self.decode(torch.randn_like(posterior_ema.sample()))
+                    log["reconstructions_ema"] = xrec_ema
+        log["inputs"] = x
+        return log
+
+    def to_rgb(self, x):
+        assert self.image_key == "segmentation"
+        if not hasattr(self, "colorize"):
+            self.register_buffer("colorize", torch.randn(3, x.shape[1], 1, 1).to(x))
+        x = F.conv2d(x, weight=self.colorize)
+        x = 2.*(x-x.min())/(x.max()-x.min()) - 1.
+        return x
+
+
+class IdentityFirstStage(torch.nn.Module):
+    def __init__(self, *args, vq_interface=False, **kwargs):
+        self.vq_interface = vq_interface
+        super().__init__()
+
+    def encode(self, x, *args, **kwargs):
+        return x
+
+    def decode(self, x, *args, **kwargs):
+        return x
+
+    def quantize(self, x, *args, **kwargs):
+        if self.vq_interface:
+            return x, None, [None, None, None]
+        return x
+
+    def forward(self, x, *args, **kwargs):
+        return x
+

ldm/models/diffusion/ddim.py

@@ -0,0 +1,336 @@
+"""SAMPLING ONLY."""
+
+import torch
+import numpy as np
+from tqdm import tqdm
+
+from ldm.modules.diffusionmodules.util import make_ddim_sampling_parameters, make_ddim_timesteps, noise_like, extract_into_tensor
+
+
+class DDIMSampler(object):
+    def __init__(self, model, schedule="linear", **kwargs):
+        super().__init__()
+        self.model = model
+        self.ddpm_num_timesteps = model.num_timesteps
+        self.schedule = schedule
+
+    def register_buffer(self, name, attr):
+        if type(attr) == torch.Tensor:
+            if attr.device != torch.device("cuda"):
+                attr = attr.to(torch.device("cuda"))
+        setattr(self, name, attr)
+
+    def make_schedule(self, ddim_num_steps, ddim_discretize="uniform", ddim_eta=0., verbose=True):
+        self.ddim_timesteps = make_ddim_timesteps(ddim_discr_method=ddim_discretize, num_ddim_timesteps=ddim_num_steps,
+                                                  num_ddpm_timesteps=self.ddpm_num_timesteps,verbose=verbose)
+        alphas_cumprod = self.model.alphas_cumprod
+        assert alphas_cumprod.shape[0] == self.ddpm_num_timesteps, 'alphas have to be defined for each timestep'
+        to_torch = lambda x: x.clone().detach().to(torch.float32).to(self.model.device)
+
+        self.register_buffer('betas', to_torch(self.model.betas))
+        self.register_buffer('alphas_cumprod', to_torch(alphas_cumprod))
+        self.register_buffer('alphas_cumprod_prev', to_torch(self.model.alphas_cumprod_prev))
+
+        # calculations for diffusion q(x_t | x_{t-1}) and others
+        self.register_buffer('sqrt_alphas_cumprod', to_torch(np.sqrt(alphas_cumprod.cpu())))
+        self.register_buffer('sqrt_one_minus_alphas_cumprod', to_torch(np.sqrt(1. - alphas_cumprod.cpu())))
+        self.register_buffer('log_one_minus_alphas_cumprod', to_torch(np.log(1. - alphas_cumprod.cpu())))
+        self.register_buffer('sqrt_recip_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod.cpu())))
+        self.register_buffer('sqrt_recipm1_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod.cpu() - 1)))
+
+        # ddim sampling parameters
+        ddim_sigmas, ddim_alphas, ddim_alphas_prev = make_ddim_sampling_parameters(alphacums=alphas_cumprod.cpu(),
+                                                                                   ddim_timesteps=self.ddim_timesteps,
+                                                                                   eta=ddim_eta,verbose=verbose)
+        self.register_buffer('ddim_sigmas', ddim_sigmas)
+        self.register_buffer('ddim_alphas', ddim_alphas)
+        self.register_buffer('ddim_alphas_prev', ddim_alphas_prev)
+        self.register_buffer('ddim_sqrt_one_minus_alphas', np.sqrt(1. - ddim_alphas))
+        sigmas_for_original_sampling_steps = ddim_eta * torch.sqrt(
+            (1 - self.alphas_cumprod_prev) / (1 - self.alphas_cumprod) * (
+                        1 - self.alphas_cumprod / self.alphas_cumprod_prev))
+        self.register_buffer('ddim_sigmas_for_original_num_steps', sigmas_for_original_sampling_steps)
+
+    @torch.no_grad()
+    def sample(self,
+               S,
+               batch_size,
+               shape,
+               conditioning=None,
+               callback=None,
+               normals_sequence=None,
+               img_callback=None,
+               quantize_x0=False,
+               eta=0.,
+               mask=None,
+               x0=None,
+               temperature=1.,
+               noise_dropout=0.,
+               score_corrector=None,
+               corrector_kwargs=None,
+               verbose=True,
+               x_T=None,
+               log_every_t=100,
+               unconditional_guidance_scale=1.,
+               unconditional_conditioning=None, # this has to come in the same format as the conditioning, # e.g. as encoded tokens, ...
+               dynamic_threshold=None,
+               ucg_schedule=None,
+               **kwargs
+               ):
+        if conditioning is not None:
+            if isinstance(conditioning, dict):
+                ctmp = conditioning[list(conditioning.keys())[0]]
+                while isinstance(ctmp, list): ctmp = ctmp[0]
+                cbs = ctmp.shape[0]
+                if cbs != batch_size:
+                    print(f"Warning: Got {cbs} conditionings but batch-size is {batch_size}")
+
+            elif isinstance(conditioning, list):
+                for ctmp in conditioning:
+                    if ctmp.shape[0] != batch_size:
+                        print(f"Warning: Got {cbs} conditionings but batch-size is {batch_size}")
+
+            else:
+                if conditioning.shape[0] != batch_size:
+                    print(f"Warning: Got {conditioning.shape[0]} conditionings but batch-size is {batch_size}")
+
+        self.make_schedule(ddim_num_steps=S, ddim_eta=eta, verbose=verbose)
+        # sampling
+        C, H, W = shape
+        size = (batch_size, C, H, W)
+        print(f'Data shape for DDIM sampling is {size}, eta {eta}')
+
+        samples, intermediates = self.ddim_sampling(conditioning, size,
+                                                    callback=callback,
+                                                    img_callback=img_callback,
+                                                    quantize_denoised=quantize_x0,
+                                                    mask=mask, x0=x0,
+                                                    ddim_use_original_steps=False,
+                                                    noise_dropout=noise_dropout,
+                                                    temperature=temperature,
+                                                    score_corrector=score_corrector,
+                                                    corrector_kwargs=corrector_kwargs,
+                                                    x_T=x_T,
+                                                    log_every_t=log_every_t,
+                                                    unconditional_guidance_scale=unconditional_guidance_scale,
+                                                    unconditional_conditioning=unconditional_conditioning,
+                                                    dynamic_threshold=dynamic_threshold,
+                                                    ucg_schedule=ucg_schedule
+                                                    )
+        return samples, intermediates
+
+    @torch.no_grad()
+    def ddim_sampling(self, cond, shape,
+                      x_T=None, ddim_use_original_steps=False,
+                      callback=None, timesteps=None, quantize_denoised=False,
+                      mask=None, x0=None, img_callback=None, log_every_t=100,
+                      temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None,
+                      unconditional_guidance_scale=1., unconditional_conditioning=None, dynamic_threshold=None,
+                      ucg_schedule=None):
+        device = self.model.betas.device
+        b = shape[0]
+        if x_T is None:
+            img = torch.randn(shape, device=device)
+        else:
+            img = x_T
+
+        if timesteps is None:
+            timesteps = self.ddpm_num_timesteps if ddim_use_original_steps else self.ddim_timesteps
+        elif timesteps is not None and not ddim_use_original_steps:
+            subset_end = int(min(timesteps / self.ddim_timesteps.shape[0], 1) * self.ddim_timesteps.shape[0]) - 1
+            timesteps = self.ddim_timesteps[:subset_end]
+
+        intermediates = {'x_inter': [img], 'pred_x0': [img]}
+        time_range = reversed(range(0,timesteps)) if ddim_use_original_steps else np.flip(timesteps)
+        total_steps = timesteps if ddim_use_original_steps else timesteps.shape[0]
+        print(f"Running DDIM Sampling with {total_steps} timesteps")
+
+        iterator = tqdm(time_range, desc='DDIM Sampler', total=total_steps)
+
+        for i, step in enumerate(iterator):
+            index = total_steps - i - 1
+            ts = torch.full((b,), step, device=device, dtype=torch.long)
+
+            if mask is not None:
+                assert x0 is not None
+                img_orig = self.model.q_sample(x0, ts)  # TODO: deterministic forward pass?
+                img = img_orig * mask + (1. - mask) * img
+
+            if ucg_schedule is not None:
+                assert len(ucg_schedule) == len(time_range)
+                unconditional_guidance_scale = ucg_schedule[i]
+
+            outs = self.p_sample_ddim(img, cond, ts, index=index, use_original_steps=ddim_use_original_steps,
+                                      quantize_denoised=quantize_denoised, temperature=temperature,
+                                      noise_dropout=noise_dropout, score_corrector=score_corrector,
+                                      corrector_kwargs=corrector_kwargs,
+                                      unconditional_guidance_scale=unconditional_guidance_scale,
+                                      unconditional_conditioning=unconditional_conditioning,
+                                      dynamic_threshold=dynamic_threshold)
+            img, pred_x0 = outs
+            if callback: callback(i)
+            if img_callback: img_callback(pred_x0, i)
+
+            if index % log_every_t == 0 or index == total_steps - 1:
+                intermediates['x_inter'].append(img)
+                intermediates['pred_x0'].append(pred_x0)
+
+        return img, intermediates
+
+    @torch.no_grad()
+    def p_sample_ddim(self, x, c, t, index, repeat_noise=False, use_original_steps=False, quantize_denoised=False,
+                      temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None,
+                      unconditional_guidance_scale=1., unconditional_conditioning=None,
+                      dynamic_threshold=None):
+        b, *_, device = *x.shape, x.device
+
+        if unconditional_conditioning is None or unconditional_guidance_scale == 1.:
+            model_output = self.model.apply_model(x, t, c)
+        else:
+            x_in = torch.cat([x] * 2)
+            t_in = torch.cat([t] * 2)
+            if isinstance(c, dict):
+                assert isinstance(unconditional_conditioning, dict)
+                c_in = dict()
+                for k in c:
+                    if isinstance(c[k], list):
+                        c_in[k] = [torch.cat([
+                            unconditional_conditioning[k][i],
+                            c[k][i]]) for i in range(len(c[k]))]
+                    else:
+                        c_in[k] = torch.cat([
+                                unconditional_conditioning[k],
+                                c[k]])
+            elif isinstance(c, list):
+                c_in = list()
+                assert isinstance(unconditional_conditioning, list)
+                for i in range(len(c)):
+                    c_in.append(torch.cat([unconditional_conditioning[i], c[i]]))
+            else:
+                c_in = torch.cat([unconditional_conditioning, c])
+            model_uncond, model_t = self.model.apply_model(x_in, t_in, c_in).chunk(2)
+            model_output = model_uncond + unconditional_guidance_scale * (model_t - model_uncond)
+
+        if self.model.parameterization == "v":
+            e_t = self.model.predict_eps_from_z_and_v(x, t, model_output)
+        else:
+            e_t = model_output
+
+        if score_corrector is not None:
+            assert self.model.parameterization == "eps", 'not implemented'
+            e_t = score_corrector.modify_score(self.model, e_t, x, t, c, **corrector_kwargs)
+
+        alphas = self.model.alphas_cumprod if use_original_steps else self.ddim_alphas
+        alphas_prev = self.model.alphas_cumprod_prev if use_original_steps else self.ddim_alphas_prev
+        sqrt_one_minus_alphas = self.model.sqrt_one_minus_alphas_cumprod if use_original_steps else self.ddim_sqrt_one_minus_alphas
+        sigmas = self.model.ddim_sigmas_for_original_num_steps if use_original_steps else self.ddim_sigmas
+        # select parameters corresponding to the currently considered timestep
+        a_t = torch.full((b, 1, 1, 1), alphas[index], device=device)
+        a_prev = torch.full((b, 1, 1, 1), alphas_prev[index], device=device)
+        sigma_t = torch.full((b, 1, 1, 1), sigmas[index], device=device)
+        sqrt_one_minus_at = torch.full((b, 1, 1, 1), sqrt_one_minus_alphas[index],device=device)
+
+        # current prediction for x_0
+        if self.model.parameterization != "v":
+            pred_x0 = (x - sqrt_one_minus_at * e_t) / a_t.sqrt()
+        else:
+            pred_x0 = self.model.predict_start_from_z_and_v(x, t, model_output)
+
+        if quantize_denoised:
+            pred_x0, _, *_ = self.model.first_stage_model.quantize(pred_x0)
+
+        if dynamic_threshold is not None:
+            raise NotImplementedError()
+
+        # direction pointing to x_t
+        dir_xt = (1. - a_prev - sigma_t**2).sqrt() * e_t
+        noise = sigma_t * noise_like(x.shape, device, repeat_noise) * temperature
+        if noise_dropout > 0.:
+            noise = torch.nn.functional.dropout(noise, p=noise_dropout)
+        x_prev = a_prev.sqrt() * pred_x0 + dir_xt + noise
+        return x_prev, pred_x0
+
+    @torch.no_grad()
+    def encode(self, x0, c, t_enc, use_original_steps=False, return_intermediates=None,
+               unconditional_guidance_scale=1.0, unconditional_conditioning=None, callback=None):
+        num_reference_steps = self.ddpm_num_timesteps if use_original_steps else self.ddim_timesteps.shape[0]
+
+        assert t_enc <= num_reference_steps
+        num_steps = t_enc
+
+        if use_original_steps:
+            alphas_next = self.alphas_cumprod[:num_steps]
+            alphas = self.alphas_cumprod_prev[:num_steps]
+        else:
+            alphas_next = self.ddim_alphas[:num_steps]
+            alphas = torch.tensor(self.ddim_alphas_prev[:num_steps])
+
+        x_next = x0
+        intermediates = []
+        inter_steps = []
+        for i in tqdm(range(num_steps), desc='Encoding Image'):
+            t = torch.full((x0.shape[0],), i, device=self.model.device, dtype=torch.long)
+            if unconditional_guidance_scale == 1.:
+                noise_pred = self.model.apply_model(x_next, t, c)
+            else:
+                assert unconditional_conditioning is not None
+                e_t_uncond, noise_pred = torch.chunk(
+                    self.model.apply_model(torch.cat((x_next, x_next)), torch.cat((t, t)),
+                                           torch.cat((unconditional_conditioning, c))), 2)
+                noise_pred = e_t_uncond + unconditional_guidance_scale * (noise_pred - e_t_uncond)
+
+            xt_weighted = (alphas_next[i] / alphas[i]).sqrt() * x_next
+            weighted_noise_pred = alphas_next[i].sqrt() * (
+                    (1 / alphas_next[i] - 1).sqrt() - (1 / alphas[i] - 1).sqrt()) * noise_pred
+            x_next = xt_weighted + weighted_noise_pred
+            if return_intermediates and i % (
+                    num_steps // return_intermediates) == 0 and i < num_steps - 1:
+                intermediates.append(x_next)
+                inter_steps.append(i)
+            elif return_intermediates and i >= num_steps - 2:
+                intermediates.append(x_next)
+                inter_steps.append(i)
+            if callback: callback(i)
+
+        out = {'x_encoded': x_next, 'intermediate_steps': inter_steps}
+        if return_intermediates:
+            out.update({'intermediates': intermediates})
+        return x_next, out
+
+    @torch.no_grad()
+    def stochastic_encode(self, x0, t, use_original_steps=False, noise=None):
+        # fast, but does not allow for exact reconstruction
+        # t serves as an index to gather the correct alphas
+        if use_original_steps:
+            sqrt_alphas_cumprod = self.sqrt_alphas_cumprod
+            sqrt_one_minus_alphas_cumprod = self.sqrt_one_minus_alphas_cumprod
+        else:
+            sqrt_alphas_cumprod = torch.sqrt(self.ddim_alphas)
+            sqrt_one_minus_alphas_cumprod = self.ddim_sqrt_one_minus_alphas
+
+        if noise is None:
+            noise = torch.randn_like(x0)
+        return (extract_into_tensor(sqrt_alphas_cumprod, t, x0.shape) * x0 +
+                extract_into_tensor(sqrt_one_minus_alphas_cumprod, t, x0.shape) * noise)
+
+    @torch.no_grad()
+    def decode(self, x_latent, cond, t_start, unconditional_guidance_scale=1.0, unconditional_conditioning=None,
+               use_original_steps=False, callback=None):
+
+        timesteps = np.arange(self.ddpm_num_timesteps) if use_original_steps else self.ddim_timesteps
+        timesteps = timesteps[:t_start]
+
+        time_range = np.flip(timesteps)
+        total_steps = timesteps.shape[0]
+        print(f"Running DDIM Sampling with {total_steps} timesteps")
+
+        iterator = tqdm(time_range, desc='Decoding image', total=total_steps)
+        x_dec = x_latent
+        for i, step in enumerate(iterator):
+            index = total_steps - i - 1
+            ts = torch.full((x_latent.shape[0],), step, device=x_latent.device, dtype=torch.long)
+            x_dec, _ = self.p_sample_ddim(x_dec, cond, ts, index=index, use_original_steps=use_original_steps,
+                                          unconditional_guidance_scale=unconditional_guidance_scale,
+                                          unconditional_conditioning=unconditional_conditioning)
+            if callback: callback(i)
+        return x_dec

ldm/models/diffusion/ddpm.py

@@ -0,0 +1,1795 @@
+"""
+wild mixture of
+https://github.com/lucidrains/denoising-diffusion-pytorch/blob/7706bdfc6f527f58d33f84b7b522e61e6e3164b3/denoising_diffusion_pytorch/denoising_diffusion_pytorch.py
+https://github.com/openai/improved-diffusion/blob/e94489283bb876ac1477d5dd7709bbbd2d9902ce/improved_diffusion/gaussian_diffusion.py
+https://github.com/CompVis/taming-transformers
+-- merci
+"""
+
+import torch
+import torch.nn as nn
+import numpy as np
+import pytorch_lightning as pl
+from torch.optim.lr_scheduler import LambdaLR
+from einops import rearrange, repeat
+from contextlib import contextmanager, nullcontext
+from functools import partial
+import itertools
+from tqdm import tqdm
+from torchvision.utils import make_grid
+from pytorch_lightning.utilities.distributed import rank_zero_only
+from omegaconf import ListConfig
+
+from ldm.util import log_txt_as_img, exists, default, ismap, isimage, mean_flat, count_params, instantiate_from_config
+from ldm.modules.ema import LitEma
+from ldm.modules.distributions.distributions import normal_kl, DiagonalGaussianDistribution
+from ldm.models.autoencoder import IdentityFirstStage, AutoencoderKL
+from ldm.modules.diffusionmodules.util import make_beta_schedule, extract_into_tensor, noise_like
+from ldm.models.diffusion.ddim import DDIMSampler
+
+
+__conditioning_keys__ = {'concat': 'c_concat',
+                         'crossattn': 'c_crossattn',
+                         'adm': 'y'}
+
+
+def disabled_train(self, mode=True):
+    """Overwrite model.train with this function to make sure train/eval mode
+    does not change anymore."""
+    return self
+
+
+def uniform_on_device(r1, r2, shape, device):
+    return (r1 - r2) * torch.rand(*shape, device=device) + r2
+
+
+class DDPM(pl.LightningModule):
+    # classic DDPM with Gaussian diffusion, in image space
+    def __init__(self,
+                 unet_config,
+                 timesteps=1000,
+                 beta_schedule="linear",
+                 loss_type="l2",
+                 ckpt_path=None,
+                 ignore_keys=[],
+                 load_only_unet=False,
+                 monitor="val/loss",
+                 use_ema=True,
+                 first_stage_key="image",
+                 image_size=256,
+                 channels=3,
+                 log_every_t=100,
+                 clip_denoised=True,
+                 linear_start=1e-4,
+                 linear_end=2e-2,
+                 cosine_s=8e-3,
+                 given_betas=None,
+                 original_elbo_weight=0.,
+                 v_posterior=0.,  # weight for choosing posterior variance as sigma = (1-v) * beta_tilde + v * beta
+                 l_simple_weight=1.,
+                 conditioning_key=None,
+                 parameterization="eps",  # all assuming fixed variance schedules
+                 scheduler_config=None,
+                 use_positional_encodings=False,
+                 learn_logvar=False,
+                 logvar_init=0.,
+                 make_it_fit=False,
+                 ucg_training=None,
+                 reset_ema=False,
+                 reset_num_ema_updates=False,
+                 ):
+        super().__init__()
+        assert parameterization in ["eps", "x0", "v"], 'currently only supporting "eps" and "x0" and "v"'
+        self.parameterization = parameterization
+        print(f"{self.__class__.__name__}: Running in {self.parameterization}-prediction mode")
+        self.cond_stage_model = None
+        self.clip_denoised = clip_denoised
+        self.log_every_t = log_every_t
+        self.first_stage_key = first_stage_key
+        self.image_size = image_size  # try conv?
+        self.channels = channels
+        self.use_positional_encodings = use_positional_encodings
+        self.model = DiffusionWrapper(unet_config, conditioning_key)
+        count_params(self.model, verbose=True)
+        self.use_ema = use_ema
+        if self.use_ema:
+            self.model_ema = LitEma(self.model)
+            print(f"Keeping EMAs of {len(list(self.model_ema.buffers()))}.")
+
+        self.use_scheduler = scheduler_config is not None
+        if self.use_scheduler:
+            self.scheduler_config = scheduler_config
+
+        self.v_posterior = v_posterior
+        self.original_elbo_weight = original_elbo_weight
+        self.l_simple_weight = l_simple_weight
+
+        if monitor is not None:
+            self.monitor = monitor
+        self.make_it_fit = make_it_fit
+        if reset_ema: assert exists(ckpt_path)
+        if ckpt_path is not None:
+            self.init_from_ckpt(ckpt_path, ignore_keys=ignore_keys, only_model=load_only_unet)
+            if reset_ema:
+                assert self.use_ema
+                print(f"Resetting ema to pure model weights. This is useful when restoring from an ema-only checkpoint.")
+                self.model_ema = LitEma(self.model)
+        if reset_num_ema_updates:
+            print(" +++++++++++ WARNING: RESETTING NUM_EMA UPDATES TO ZERO +++++++++++ ")
+            assert self.use_ema
+            self.model_ema.reset_num_updates()
+
+        self.register_schedule(given_betas=given_betas, beta_schedule=beta_schedule, timesteps=timesteps,
+                               linear_start=linear_start, linear_end=linear_end, cosine_s=cosine_s)
+
+        self.loss_type = loss_type
+
+        self.learn_logvar = learn_logvar
+        self.logvar = torch.full(fill_value=logvar_init, size=(self.num_timesteps,))
+        if self.learn_logvar:
+            self.logvar = nn.Parameter(self.logvar, requires_grad=True)
+
+        self.ucg_training = ucg_training or dict()
+        if self.ucg_training:
+            self.ucg_prng = np.random.RandomState()
+
+    def register_schedule(self, given_betas=None, beta_schedule="linear", timesteps=1000,
+                          linear_start=1e-4, linear_end=2e-2, cosine_s=8e-3):
+        if exists(given_betas):
+            betas = given_betas
+        else:
+            betas = make_beta_schedule(beta_schedule, timesteps, linear_start=linear_start, linear_end=linear_end,
+                                       cosine_s=cosine_s)
+        alphas = 1. - betas
+        alphas_cumprod = np.cumprod(alphas, axis=0)
+        alphas_cumprod_prev = np.append(1., alphas_cumprod[:-1])
+
+        timesteps, = betas.shape
+        self.num_timesteps = int(timesteps)
+        self.linear_start = linear_start
+        self.linear_end = linear_end
+        assert alphas_cumprod.shape[0] == self.num_timesteps, 'alphas have to be defined for each timestep'
+
+        to_torch = partial(torch.tensor, dtype=torch.float32)
+
+        self.register_buffer('betas', to_torch(betas))
+        self.register_buffer('alphas_cumprod', to_torch(alphas_cumprod))
+        self.register_buffer('alphas_cumprod_prev', to_torch(alphas_cumprod_prev))
+
+        # calculations for diffusion q(x_t | x_{t-1}) and others
+        self.register_buffer('sqrt_alphas_cumprod', to_torch(np.sqrt(alphas_cumprod)))
+        self.register_buffer('sqrt_one_minus_alphas_cumprod', to_torch(np.sqrt(1. - alphas_cumprod)))
+        self.register_buffer('log_one_minus_alphas_cumprod', to_torch(np.log(1. - alphas_cumprod)))
+        self.register_buffer('sqrt_recip_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod)))
+        self.register_buffer('sqrt_recipm1_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod - 1)))
+
+        # calculations for posterior q(x_{t-1} | x_t, x_0)
+        posterior_variance = (1 - self.v_posterior) * betas * (1. - alphas_cumprod_prev) / (
+                1. - alphas_cumprod) + self.v_posterior * betas
+        # above: equal to 1. / (1. / (1. - alpha_cumprod_tm1) + alpha_t / beta_t)
+        self.register_buffer('posterior_variance', to_torch(posterior_variance))
+        # below: log calculation clipped because the posterior variance is 0 at the beginning of the diffusion chain
+        self.register_buffer('posterior_log_variance_clipped', to_torch(np.log(np.maximum(posterior_variance, 1e-20))))
+        self.register_buffer('posterior_mean_coef1', to_torch(
+            betas * np.sqrt(alphas_cumprod_prev) / (1. - alphas_cumprod)))
+        self.register_buffer('posterior_mean_coef2', to_torch(
+            (1. - alphas_cumprod_prev) * np.sqrt(alphas) / (1. - alphas_cumprod)))
+
+        if self.parameterization == "eps":
+            lvlb_weights = self.betas ** 2 / (
+                    2 * self.posterior_variance * to_torch(alphas) * (1 - self.alphas_cumprod))
+        elif self.parameterization == "x0":
+            lvlb_weights = 0.5 * np.sqrt(torch.Tensor(alphas_cumprod)) / (2. * 1 - torch.Tensor(alphas_cumprod))
+        elif self.parameterization == "v":
+            lvlb_weights = torch.ones_like(self.betas ** 2 / (
+                    2 * self.posterior_variance * to_torch(alphas) * (1 - self.alphas_cumprod)))
+        else:
+            raise NotImplementedError("mu not supported")
+        lvlb_weights[0] = lvlb_weights[1]
+        self.register_buffer('lvlb_weights', lvlb_weights, persistent=False)
+        assert not torch.isnan(self.lvlb_weights).all()
+
+    @contextmanager
+    def ema_scope(self, context=None):
+        if self.use_ema:
+            self.model_ema.store(self.model.parameters())
+            self.model_ema.copy_to(self.model)
+            if context is not None:
+                print(f"{context}: Switched to EMA weights")
+        try:
+            yield None
+        finally:
+            if self.use_ema:
+                self.model_ema.restore(self.model.parameters())
+                if context is not None:
+                    print(f"{context}: Restored training weights")
+
+    @torch.no_grad()
+    def init_from_ckpt(self, path, ignore_keys=list(), only_model=False):
+        sd = torch.load(path, map_location="cpu")
+        if "state_dict" in list(sd.keys()):
+            sd = sd["state_dict"]
+        keys = list(sd.keys())
+        for k in keys:
+            for ik in ignore_keys:
+                if k.startswith(ik):
+                    print("Deleting key {} from state_dict.".format(k))
+                    del sd[k]
+        if self.make_it_fit:
+            n_params = len([name for name, _ in
+                            itertools.chain(self.named_parameters(),
+                                            self.named_buffers())])
+            for name, param in tqdm(
+                    itertools.chain(self.named_parameters(),
+                                    self.named_buffers()),
+                    desc="Fitting old weights to new weights",
+                    total=n_params
+            ):
+                if not name in sd:
+                    continue
+                old_shape = sd[name].shape
+                new_shape = param.shape
+                assert len(old_shape) == len(new_shape)
+                if len(new_shape) > 2:
+                    # we only modify first two axes
+                    assert new_shape[2:] == old_shape[2:]
+                # assumes first axis corresponds to output dim
+                if not new_shape == old_shape:
+                    new_param = param.clone()
+                    old_param = sd[name]
+                    if len(new_shape) == 1:
+                        for i in range(new_param.shape[0]):
+                            new_param[i] = old_param[i % old_shape[0]]
+                    elif len(new_shape) >= 2:
+                        for i in range(new_param.shape[0]):
+                            for j in range(new_param.shape[1]):
+                                new_param[i, j] = old_param[i % old_shape[0], j % old_shape[1]]
+
+                        n_used_old = torch.ones(old_shape[1])
+                        for j in range(new_param.shape[1]):
+                            n_used_old[j % old_shape[1]] += 1
+                        n_used_new = torch.zeros(new_shape[1])
+                        for j in range(new_param.shape[1]):
+                            n_used_new[j] = n_used_old[j % old_shape[1]]
+
+                        n_used_new = n_used_new[None, :]
+                        while len(n_used_new.shape) < len(new_shape):
+                            n_used_new = n_used_new.unsqueeze(-1)
+                        new_param /= n_used_new
+
+                    sd[name] = new_param
+
+        missing, unexpected = self.load_state_dict(sd, strict=False) if not only_model else self.model.load_state_dict(
+            sd, strict=False)
+        print(f"Restored from {path} with {len(missing)} missing and {len(unexpected)} unexpected keys")
+        if len(missing) > 0:
+            print(f"Missing Keys:\n {missing}")
+        if len(unexpected) > 0:
+            print(f"\nUnexpected Keys:\n {unexpected}")
+
+    def q_mean_variance(self, x_start, t):
+        """
+        Get the distribution q(x_t | x_0).
+        :param x_start: the [N x C x ...] tensor of noiseless inputs.
+        :param t: the number of diffusion steps (minus 1). Here, 0 means one step.
+        :return: A tuple (mean, variance, log_variance), all of x_start's shape.
+        """
+        mean = (extract_into_tensor(self.sqrt_alphas_cumprod, t, x_start.shape) * x_start)
+        variance = extract_into_tensor(1.0 - self.alphas_cumprod, t, x_start.shape)
+        log_variance = extract_into_tensor(self.log_one_minus_alphas_cumprod, t, x_start.shape)
+        return mean, variance, log_variance
+
+    def predict_start_from_noise(self, x_t, t, noise):
+        return (
+                extract_into_tensor(self.sqrt_recip_alphas_cumprod, t, x_t.shape) * x_t -
+                extract_into_tensor(self.sqrt_recipm1_alphas_cumprod, t, x_t.shape) * noise
+        )
+
+    def predict_start_from_z_and_v(self, x_t, t, v):
+        # self.register_buffer('sqrt_alphas_cumprod', to_torch(np.sqrt(alphas_cumprod)))
+        # self.register_buffer('sqrt_one_minus_alphas_cumprod', to_torch(np.sqrt(1. - alphas_cumprod)))
+        return (
+                extract_into_tensor(self.sqrt_alphas_cumprod, t, x_t.shape) * x_t -
+                extract_into_tensor(self.sqrt_one_minus_alphas_cumprod, t, x_t.shape) * v
+        )
+
+    def predict_eps_from_z_and_v(self, x_t, t, v):
+        return (
+                extract_into_tensor(self.sqrt_alphas_cumprod, t, x_t.shape) * v +
+                extract_into_tensor(self.sqrt_one_minus_alphas_cumprod, t, x_t.shape) * x_t
+        )
+
+    def q_posterior(self, x_start, x_t, t):
+        posterior_mean = (
+                extract_into_tensor(self.posterior_mean_coef1, t, x_t.shape) * x_start +
+                extract_into_tensor(self.posterior_mean_coef2, t, x_t.shape) * x_t
+        )
+        posterior_variance = extract_into_tensor(self.posterior_variance, t, x_t.shape)
+        posterior_log_variance_clipped = extract_into_tensor(self.posterior_log_variance_clipped, t, x_t.shape)
+        return posterior_mean, posterior_variance, posterior_log_variance_clipped
+
+    def p_mean_variance(self, x, t, clip_denoised: bool):
+        model_out = self.model(x, t)
+        if self.parameterization == "eps":
+            x_recon = self.predict_start_from_noise(x, t=t, noise=model_out)
+        elif self.parameterization == "x0":
+            x_recon = model_out
+        if clip_denoised:
+            x_recon.clamp_(-1., 1.)
+
+        model_mean, posterior_variance, posterior_log_variance = self.q_posterior(x_start=x_recon, x_t=x, t=t)
+        return model_mean, posterior_variance, posterior_log_variance
+
+    @torch.no_grad()
+    def p_sample(self, x, t, clip_denoised=True, repeat_noise=False):
+        b, *_, device = *x.shape, x.device
+        model_mean, _, model_log_variance = self.p_mean_variance(x=x, t=t, clip_denoised=clip_denoised)
+        noise = noise_like(x.shape, device, repeat_noise)
+        # no noise when t == 0
+        nonzero_mask = (1 - (t == 0).float()).reshape(b, *((1,) * (len(x.shape) - 1)))
+        return model_mean + nonzero_mask * (0.5 * model_log_variance).exp() * noise
+
+    @torch.no_grad()
+    def p_sample_loop(self, shape, return_intermediates=False):
+        device = self.betas.device
+        b = shape[0]
+        img = torch.randn(shape, device=device)
+        intermediates = [img]
+        for i in tqdm(reversed(range(0, self.num_timesteps)), desc='Sampling t', total=self.num_timesteps):
+            img = self.p_sample(img, torch.full((b,), i, device=device, dtype=torch.long),
+                                clip_denoised=self.clip_denoised)
+            if i % self.log_every_t == 0 or i == self.num_timesteps - 1:
+                intermediates.append(img)
+        if return_intermediates:
+            return img, intermediates
+        return img
+
+    @torch.no_grad()
+    def sample(self, batch_size=16, return_intermediates=False):
+        image_size = self.image_size
+        channels = self.channels
+        return self.p_sample_loop((batch_size, channels, image_size, image_size),
+                                  return_intermediates=return_intermediates)
+
+    def q_sample(self, x_start, t, noise=None):
+        noise = default(noise, lambda: torch.randn_like(x_start))
+        return (extract_into_tensor(self.sqrt_alphas_cumprod, t, x_start.shape) * x_start +
+                extract_into_tensor(self.sqrt_one_minus_alphas_cumprod, t, x_start.shape) * noise)
+
+    def get_v(self, x, noise, t):
+        return (
+                extract_into_tensor(self.sqrt_alphas_cumprod, t, x.shape) * noise -
+                extract_into_tensor(self.sqrt_one_minus_alphas_cumprod, t, x.shape) * x
+        )
+
+    def get_loss(self, pred, target, mean=True):
+        if self.loss_type == 'l1':
+            loss = (target - pred).abs()
+            if mean:
+                loss = loss.mean()
+        elif self.loss_type == 'l2':
+            if mean:
+                loss = torch.nn.functional.mse_loss(target, pred)
+            else:
+                loss = torch.nn.functional.mse_loss(target, pred, reduction='none')
+        else:
+            raise NotImplementedError("unknown loss type '{loss_type}'")
+
+        return loss
+
+    def p_losses(self, x_start, t, noise=None):
+        noise = default(noise, lambda: torch.randn_like(x_start))
+        x_noisy = self.q_sample(x_start=x_start, t=t, noise=noise)
+        model_out = self.model(x_noisy, t)
+
+        loss_dict = {}
+        if self.parameterization == "eps":
+            target = noise
+        elif self.parameterization == "x0":
+            target = x_start
+        elif self.parameterization == "v":
+            target = self.get_v(x_start, noise, t)
+        else:
+            raise NotImplementedError(f"Parameterization {self.parameterization} not yet supported")
+
+        loss = self.get_loss(model_out, target, mean=False).mean(dim=[1, 2, 3])
+
+        log_prefix = 'train' if self.training else 'val'
+
+        loss_dict.update({f'{log_prefix}/loss_simple': loss.mean()})
+        loss_simple = loss.mean() * self.l_simple_weight
+
+        loss_vlb = (self.lvlb_weights[t] * loss).mean()
+        loss_dict.update({f'{log_prefix}/loss_vlb': loss_vlb})
+
+        loss = loss_simple + self.original_elbo_weight * loss_vlb
+
+        loss_dict.update({f'{log_prefix}/loss': loss})
+
+        return loss, loss_dict
+
+    def forward(self, x, *args, **kwargs):
+        # b, c, h, w, device, img_size, = *x.shape, x.device, self.image_size
+        # assert h == img_size and w == img_size, f'height and width of image must be {img_size}'
+        t = torch.randint(0, self.num_timesteps, (x.shape[0],), device=self.device).long()
+        return self.p_losses(x, t, *args, **kwargs)
+
+    def get_input(self, batch, k):
+        x = batch[k]
+        if len(x.shape) == 3:
+            x = x[..., None]
+        x = rearrange(x, 'b h w c -> b c h w')
+        x = x.to(memory_format=torch.contiguous_format).float()
+        return x
+
+    def shared_step(self, batch):
+        x = self.get_input(batch, self.first_stage_key)
+        loss, loss_dict = self(x)
+        return loss, loss_dict
+
+    def training_step(self, batch, batch_idx):
+        for k in self.ucg_training:
+            p = self.ucg_training[k]["p"]
+            val = self.ucg_training[k]["val"]
+            if val is None:
+                val = ""
+            for i in range(len(batch[k])):
+                if self.ucg_prng.choice(2, p=[1 - p, p]):
+                    batch[k][i] = val
+
+        loss, loss_dict = self.shared_step(batch)
+
+        self.log_dict(loss_dict, prog_bar=True,
+                      logger=True, on_step=True, on_epoch=True)
+
+        self.log("global_step", self.global_step,
+                 prog_bar=True, logger=True, on_step=True, on_epoch=False)
+
+        if self.use_scheduler:
+            lr = self.optimizers().param_groups[0]['lr']
+            self.log('lr_abs', lr, prog_bar=True, logger=True, on_step=True, on_epoch=False)
+
+        return loss
+
+    @torch.no_grad()
+    def validation_step(self, batch, batch_idx):
+        _, loss_dict_no_ema = self.shared_step(batch)
+        with self.ema_scope():
+            _, loss_dict_ema = self.shared_step(batch)
+            loss_dict_ema = {key + '_ema': loss_dict_ema[key] for key in loss_dict_ema}
+        self.log_dict(loss_dict_no_ema, prog_bar=False, logger=True, on_step=False, on_epoch=True)
+        self.log_dict(loss_dict_ema, prog_bar=False, logger=True, on_step=False, on_epoch=True)
+
+    def on_train_batch_end(self, *args, **kwargs):
+        if self.use_ema:
+            self.model_ema(self.model)
+
+    def _get_rows_from_list(self, samples):
+        n_imgs_per_row = len(samples)
+        denoise_grid = rearrange(samples, 'n b c h w -> b n c h w')
+        denoise_grid = rearrange(denoise_grid, 'b n c h w -> (b n) c h w')
+        denoise_grid = make_grid(denoise_grid, nrow=n_imgs_per_row)
+        return denoise_grid
+
+    @torch.no_grad()
+    def log_images(self, batch, N=8, n_row=2, sample=True, return_keys=None, **kwargs):
+        log = dict()
+        x = self.get_input(batch, self.first_stage_key)
+        N = min(x.shape[0], N)
+        n_row = min(x.shape[0], n_row)
+        x = x.to(self.device)[:N]
+        log["inputs"] = x
+
+        # get diffusion row
+        diffusion_row = list()
+        x_start = x[:n_row]
+
+        for t in range(self.num_timesteps):
+            if t % self.log_every_t == 0 or t == self.num_timesteps - 1:
+                t = repeat(torch.tensor([t]), '1 -> b', b=n_row)
+                t = t.to(self.device).long()
+                noise = torch.randn_like(x_start)
+                x_noisy = self.q_sample(x_start=x_start, t=t, noise=noise)
+                diffusion_row.append(x_noisy)
+
+        log["diffusion_row"] = self._get_rows_from_list(diffusion_row)
+
+        if sample:
+            # get denoise row
+            with self.ema_scope("Plotting"):
+                samples, denoise_row = self.sample(batch_size=N, return_intermediates=True)
+
+            log["samples"] = samples
+            log["denoise_row"] = self._get_rows_from_list(denoise_row)
+
+        if return_keys:
+            if np.intersect1d(list(log.keys()), return_keys).shape[0] == 0:
+                return log
+            else:
+                return {key: log[key] for key in return_keys}
+        return log
+
+    def configure_optimizers(self):
+        lr = self.learning_rate
+        params = list(self.model.parameters())
+        if self.learn_logvar:
+            params = params + [self.logvar]
+        opt = torch.optim.AdamW(params, lr=lr)
+        return opt
+
+
+class LatentDiffusion(DDPM):
+    """main class"""
+
+    def __init__(self,
+                 first_stage_config,
+                 cond_stage_config,
+                 num_timesteps_cond=None,
+                 cond_stage_key="image",
+                 cond_stage_trainable=False,
+                 concat_mode=True,
+                 cond_stage_forward=None,
+                 conditioning_key=None,
+                 scale_factor=1.0,
+                 scale_by_std=False,
+                 force_null_conditioning=False,
+                 *args, **kwargs):
+        self.force_null_conditioning = force_null_conditioning
+        self.num_timesteps_cond = default(num_timesteps_cond, 1)
+        self.scale_by_std = scale_by_std
+        assert self.num_timesteps_cond <= kwargs['timesteps']
+        # for backwards compatibility after implementation of DiffusionWrapper
+        if conditioning_key is None:
+            conditioning_key = 'concat' if concat_mode else 'crossattn'
+        if cond_stage_config == '__is_unconditional__' and not self.force_null_conditioning:
+            conditioning_key = None
+        ckpt_path = kwargs.pop("ckpt_path", None)
+        reset_ema = kwargs.pop("reset_ema", False)
+        reset_num_ema_updates = kwargs.pop("reset_num_ema_updates", False)
+        ignore_keys = kwargs.pop("ignore_keys", [])
+        super().__init__(conditioning_key=conditioning_key, *args, **kwargs)
+        self.concat_mode = concat_mode
+        self.cond_stage_trainable = cond_stage_trainable
+        self.cond_stage_key = cond_stage_key
+        try:
+            self.num_downs = len(first_stage_config.params.ddconfig.ch_mult) - 1
+        except:
+            self.num_downs = 0
+        if not scale_by_std:
+            self.scale_factor = scale_factor
+        else:
+            self.register_buffer('scale_factor', torch.tensor(scale_factor))
+        self.instantiate_first_stage(first_stage_config)
+        self.instantiate_cond_stage(cond_stage_config)
+        self.cond_stage_forward = cond_stage_forward
+        self.clip_denoised = False
+        self.bbox_tokenizer = None
+
+        self.restarted_from_ckpt = False
+        if ckpt_path is not None:
+            self.init_from_ckpt(ckpt_path, ignore_keys)
+            self.restarted_from_ckpt = True
+            if reset_ema:
+                assert self.use_ema
+                print(
+                    f"Resetting ema to pure model weights. This is useful when restoring from an ema-only checkpoint.")
+                self.model_ema = LitEma(self.model)
+        if reset_num_ema_updates:
+            print(" +++++++++++ WARNING: RESETTING NUM_EMA UPDATES TO ZERO +++++++++++ ")
+            assert self.use_ema
+            self.model_ema.reset_num_updates()
+
+    def make_cond_schedule(self, ):
+        self.cond_ids = torch.full(size=(self.num_timesteps,), fill_value=self.num_timesteps - 1, dtype=torch.long)
+        ids = torch.round(torch.linspace(0, self.num_timesteps - 1, self.num_timesteps_cond)).long()
+        self.cond_ids[:self.num_timesteps_cond] = ids
+
+    @rank_zero_only
+    @torch.no_grad()
+    def on_train_batch_start(self, batch, batch_idx, dataloader_idx):
+        # only for very first batch
+        if self.scale_by_std and self.current_epoch == 0 and self.global_step == 0 and batch_idx == 0 and not self.restarted_from_ckpt:
+            assert self.scale_factor == 1., 'rather not use custom rescaling and std-rescaling simultaneously'
+            # set rescale weight to 1./std of encodings
+            print("### USING STD-RESCALING ###")
+            x = super().get_input(batch, self.first_stage_key)
+            x = x.to(self.device)
+            encoder_posterior = self.encode_first_stage(x)
+            z = self.get_first_stage_encoding(encoder_posterior).detach()
+            del self.scale_factor
+            self.register_buffer('scale_factor', 1. / z.flatten().std())
+            print(f"setting self.scale_factor to {self.scale_factor}")
+            print("### USING STD-RESCALING ###")
+
+    def register_schedule(self,
+                          given_betas=None, beta_schedule="linear", timesteps=1000,
+                          linear_start=1e-4, linear_end=2e-2, cosine_s=8e-3):
+        super().register_schedule(given_betas, beta_schedule, timesteps, linear_start, linear_end, cosine_s)
+
+        self.shorten_cond_schedule = self.num_timesteps_cond > 1
+        if self.shorten_cond_schedule:
+            self.make_cond_schedule()
+
+    def instantiate_first_stage(self, config):
+        model = instantiate_from_config(config)
+        self.first_stage_model = model.eval()
+        self.first_stage_model.train = disabled_train
+        for param in self.first_stage_model.parameters():
+            param.requires_grad = False
+
+    def instantiate_cond_stage(self, config):
+        if not self.cond_stage_trainable:
+            if config == "__is_first_stage__":
+                print("Using first stage also as cond stage.")
+                self.cond_stage_model = self.first_stage_model
+            elif config == "__is_unconditional__":
+                print(f"Training {self.__class__.__name__} as an unconditional model.")
+                self.cond_stage_model = None
+                # self.be_unconditional = True
+            else:
+                model = instantiate_from_config(config)
+                self.cond_stage_model = model.eval()
+                self.cond_stage_model.train = disabled_train
+                for param in self.cond_stage_model.parameters():
+                    param.requires_grad = False
+        else:
+            assert config != '__is_first_stage__'
+            assert config != '__is_unconditional__'
+            model = instantiate_from_config(config)
+            self.cond_stage_model = model
+
+    def _get_denoise_row_from_list(self, samples, desc='', force_no_decoder_quantization=False):
+        denoise_row = []
+        for zd in tqdm(samples, desc=desc):
+            denoise_row.append(self.decode_first_stage(zd.to(self.device),
+                                                       force_not_quantize=force_no_decoder_quantization))
+        n_imgs_per_row = len(denoise_row)
+        denoise_row = torch.stack(denoise_row)  # n_log_step, n_row, C, H, W
+        denoise_grid = rearrange(denoise_row, 'n b c h w -> b n c h w')
+        denoise_grid = rearrange(denoise_grid, 'b n c h w -> (b n) c h w')
+        denoise_grid = make_grid(denoise_grid, nrow=n_imgs_per_row)
+        return denoise_grid
+
+    def get_first_stage_encoding(self, encoder_posterior):
+        if isinstance(encoder_posterior, DiagonalGaussianDistribution):
+            z = encoder_posterior.sample()
+        elif isinstance(encoder_posterior, torch.Tensor):
+            z = encoder_posterior
+        else:
+            raise NotImplementedError(f"encoder_posterior of type '{type(encoder_posterior)}' not yet implemented")
+        return self.scale_factor * z
+
+    def get_learned_conditioning(self, c):
+        if self.cond_stage_forward is None:
+            if hasattr(self.cond_stage_model, 'encode') and callable(self.cond_stage_model.encode):
+                c = self.cond_stage_model.encode(c)
+                if isinstance(c, DiagonalGaussianDistribution):
+                    c = c.mode()
+            else:
+                c = self.cond_stage_model(c)
+        else:
+            assert hasattr(self.cond_stage_model, self.cond_stage_forward)
+            c = getattr(self.cond_stage_model, self.cond_stage_forward)(c)
+        return c
+
+    def meshgrid(self, h, w):
+        y = torch.arange(0, h).view(h, 1, 1).repeat(1, w, 1)
+        x = torch.arange(0, w).view(1, w, 1).repeat(h, 1, 1)
+
+        arr = torch.cat([y, x], dim=-1)
+        return arr
+
+    def delta_border(self, h, w):
+        """
+        :param h: height
+        :param w: width
+        :return: normalized distance to image border,
+         wtith min distance = 0 at border and max dist = 0.5 at image center
+        """
+        lower_right_corner = torch.tensor([h - 1, w - 1]).view(1, 1, 2)
+        arr = self.meshgrid(h, w) / lower_right_corner
+        dist_left_up = torch.min(arr, dim=-1, keepdims=True)[0]
+        dist_right_down = torch.min(1 - arr, dim=-1, keepdims=True)[0]
+        edge_dist = torch.min(torch.cat([dist_left_up, dist_right_down], dim=-1), dim=-1)[0]
+        return edge_dist
+
+    def get_weighting(self, h, w, Ly, Lx, device):
+        weighting = self.delta_border(h, w)
+        weighting = torch.clip(weighting, self.split_input_params["clip_min_weight"],
+                               self.split_input_params["clip_max_weight"], )
+        weighting = weighting.view(1, h * w, 1).repeat(1, 1, Ly * Lx).to(device)
+
+        if self.split_input_params["tie_braker"]:
+            L_weighting = self.delta_border(Ly, Lx)
+            L_weighting = torch.clip(L_weighting,
+                                     self.split_input_params["clip_min_tie_weight"],
+                                     self.split_input_params["clip_max_tie_weight"])
+
+            L_weighting = L_weighting.view(1, 1, Ly * Lx).to(device)
+            weighting = weighting * L_weighting
+        return weighting
+
+    def get_fold_unfold(self, x, kernel_size, stride, uf=1, df=1):  # todo load once not every time, shorten code
+        """
+        :param x: img of size (bs, c, h, w)
+        :return: n img crops of size (n, bs, c, kernel_size[0], kernel_size[1])
+        """
+        bs, nc, h, w = x.shape
+
+        # number of crops in image
+        Ly = (h - kernel_size[0]) // stride[0] + 1
+        Lx = (w - kernel_size[1]) // stride[1] + 1
+
+        if uf == 1 and df == 1:
+            fold_params = dict(kernel_size=kernel_size, dilation=1, padding=0, stride=stride)
+            unfold = torch.nn.Unfold(**fold_params)
+
+            fold = torch.nn.Fold(output_size=x.shape[2:], **fold_params)
+
+            weighting = self.get_weighting(kernel_size[0], kernel_size[1], Ly, Lx, x.device).to(x.dtype)
+            normalization = fold(weighting).view(1, 1, h, w)  # normalizes the overlap
+            weighting = weighting.view((1, 1, kernel_size[0], kernel_size[1], Ly * Lx))
+
+        elif uf > 1 and df == 1:
+            fold_params = dict(kernel_size=kernel_size, dilation=1, padding=0, stride=stride)
+            unfold = torch.nn.Unfold(**fold_params)
+
+            fold_params2 = dict(kernel_size=(kernel_size[0] * uf, kernel_size[0] * uf),
+                                dilation=1, padding=0,
+                                stride=(stride[0] * uf, stride[1] * uf))
+            fold = torch.nn.Fold(output_size=(x.shape[2] * uf, x.shape[3] * uf), **fold_params2)
+
+            weighting = self.get_weighting(kernel_size[0] * uf, kernel_size[1] * uf, Ly, Lx, x.device).to(x.dtype)
+            normalization = fold(weighting).view(1, 1, h * uf, w * uf)  # normalizes the overlap
+            weighting = weighting.view((1, 1, kernel_size[0] * uf, kernel_size[1] * uf, Ly * Lx))
+
+        elif df > 1 and uf == 1:
+            fold_params = dict(kernel_size=kernel_size, dilation=1, padding=0, stride=stride)
+            unfold = torch.nn.Unfold(**fold_params)
+
+            fold_params2 = dict(kernel_size=(kernel_size[0] // df, kernel_size[0] // df),
+                                dilation=1, padding=0,
+                                stride=(stride[0] // df, stride[1] // df))
+            fold = torch.nn.Fold(output_size=(x.shape[2] // df, x.shape[3] // df), **fold_params2)
+
+            weighting = self.get_weighting(kernel_size[0] // df, kernel_size[1] // df, Ly, Lx, x.device).to(x.dtype)
+            normalization = fold(weighting).view(1, 1, h // df, w // df)  # normalizes the overlap
+            weighting = weighting.view((1, 1, kernel_size[0] // df, kernel_size[1] // df, Ly * Lx))
+
+        else:
+            raise NotImplementedError
+
+        return fold, unfold, normalization, weighting
+
+    @torch.no_grad()
+    def get_input(self, batch, k, return_first_stage_outputs=False, force_c_encode=False,
+                  cond_key=None, return_original_cond=False, bs=None, return_x=False):
+        x = super().get_input(batch, k)
+        if bs is not None:
+            x = x[:bs]
+        x = x.to(self.device)
+        encoder_posterior = self.encode_first_stage(x)
+        z = self.get_first_stage_encoding(encoder_posterior).detach()
+
+        if self.model.conditioning_key is not None and not self.force_null_conditioning:
+            if cond_key is None:
+                cond_key = self.cond_stage_key
+            if cond_key != self.first_stage_key:
+                if cond_key in ['caption', 'coordinates_bbox', "txt"]:
+                    xc = batch[cond_key]
+                elif cond_key in ['class_label', 'cls']:
+                    xc = batch
+                else:
+                    xc = super().get_input(batch, cond_key).to(self.device)
+            else:
+                xc = x
+            if not self.cond_stage_trainable or force_c_encode:
+                if isinstance(xc, dict) or isinstance(xc, list):
+                    c = self.get_learned_conditioning(xc)
+                else:
+                    c = self.get_learned_conditioning(xc.to(self.device))
+            else:
+                c = xc
+            if bs is not None:
+                c = c[:bs]
+
+            if self.use_positional_encodings:
+                pos_x, pos_y = self.compute_latent_shifts(batch)
+                ckey = __conditioning_keys__[self.model.conditioning_key]
+                c = {ckey: c, 'pos_x': pos_x, 'pos_y': pos_y}
+
+        else:
+            c = None
+            xc = None
+            if self.use_positional_encodings:
+                pos_x, pos_y = self.compute_latent_shifts(batch)
+                c = {'pos_x': pos_x, 'pos_y': pos_y}
+        out = [z, c]
+        if return_first_stage_outputs:
+            xrec = self.decode_first_stage(z)
+            out.extend([x, xrec])
+        if return_x:
+            out.extend([x])
+        if return_original_cond:
+            out.append(xc)
+        return out
+
+    @torch.no_grad()
+    def decode_first_stage(self, z, predict_cids=False, force_not_quantize=False):
+        if predict_cids:
+            if z.dim() == 4:
+                z = torch.argmax(z.exp(), dim=1).long()
+            z = self.first_stage_model.quantize.get_codebook_entry(z, shape=None)
+            z = rearrange(z, 'b h w c -> b c h w').contiguous()
+
+        z = 1. / self.scale_factor * z
+        return self.first_stage_model.decode(z)
+
+    @torch.no_grad()
+    def encode_first_stage(self, x):
+        return self.first_stage_model.encode(x)
+
+    def shared_step(self, batch, **kwargs):
+        x, c = self.get_input(batch, self.first_stage_key)
+        loss = self(x, c)
+        return loss
+
+    def forward(self, x, c, *args, **kwargs):
+        t = torch.randint(0, self.num_timesteps, (x.shape[0],), device=self.device).long()
+        if self.model.conditioning_key is not None:
+            assert c is not None
+            if self.cond_stage_trainable:
+                c = self.get_learned_conditioning(c)
+            if self.shorten_cond_schedule:  # TODO: drop this option
+                tc = self.cond_ids[t].to(self.device)
+                c = self.q_sample(x_start=c, t=tc, noise=torch.randn_like(c.float()))
+        return self.p_losses(x, c, t, *args, **kwargs)
+
+    def apply_model(self, x_noisy, t, cond, return_ids=False):
+        if isinstance(cond, dict):
+            # hybrid case, cond is expected to be a dict
+            pass
+        else:
+            if not isinstance(cond, list):
+                cond = [cond]
+            key = 'c_concat' if self.model.conditioning_key == 'concat' else 'c_crossattn'
+            cond = {key: cond}
+
+        x_recon = self.model(x_noisy, t, **cond)
+
+        if isinstance(x_recon, tuple) and not return_ids:
+            return x_recon[0]
+        else:
+            return x_recon
+
+    def _predict_eps_from_xstart(self, x_t, t, pred_xstart):
+        return (extract_into_tensor(self.sqrt_recip_alphas_cumprod, t, x_t.shape) * x_t - pred_xstart) / \
+               extract_into_tensor(self.sqrt_recipm1_alphas_cumprod, t, x_t.shape)
+
+    def _prior_bpd(self, x_start):
+        """
+        Get the prior KL term for the variational lower-bound, measured in
+        bits-per-dim.
+        This term can't be optimized, as it only depends on the encoder.
+        :param x_start: the [N x C x ...] tensor of inputs.
+        :return: a batch of [N] KL values (in bits), one per batch element.
+        """
+        batch_size = x_start.shape[0]
+        t = torch.tensor([self.num_timesteps - 1] * batch_size, device=x_start.device)
+        qt_mean, _, qt_log_variance = self.q_mean_variance(x_start, t)
+        kl_prior = normal_kl(mean1=qt_mean, logvar1=qt_log_variance, mean2=0.0, logvar2=0.0)
+        return mean_flat(kl_prior) / np.log(2.0)
+
+    def p_losses(self, x_start, cond, t, noise=None):
+        noise = default(noise, lambda: torch.randn_like(x_start))
+        x_noisy = self.q_sample(x_start=x_start, t=t, noise=noise)
+        model_output = self.apply_model(x_noisy, t, cond)
+
+        loss_dict = {}
+        prefix = 'train' if self.training else 'val'
+
+        if self.parameterization == "x0":
+            target = x_start
+        elif self.parameterization == "eps":
+            target = noise
+        elif self.parameterization == "v":
+            target = self.get_v(x_start, noise, t)
+        else:
+            raise NotImplementedError()
+
+        loss_simple = self.get_loss(model_output, target, mean=False).mean([1, 2, 3])
+        loss_dict.update({f'{prefix}/loss_simple': loss_simple.mean()})
+
+        logvar_t = self.logvar[t].to(self.device)
+        loss = loss_simple / torch.exp(logvar_t) + logvar_t
+        # loss = loss_simple / torch.exp(self.logvar) + self.logvar
+        if self.learn_logvar:
+            loss_dict.update({f'{prefix}/loss_gamma': loss.mean()})
+            loss_dict.update({'logvar': self.logvar.data.mean()})
+
+        loss = self.l_simple_weight * loss.mean()
+
+        loss_vlb = self.get_loss(model_output, target, mean=False).mean(dim=(1, 2, 3))
+        loss_vlb = (self.lvlb_weights[t] * loss_vlb).mean()
+        loss_dict.update({f'{prefix}/loss_vlb': loss_vlb})
+        loss += (self.original_elbo_weight * loss_vlb)
+        loss_dict.update({f'{prefix}/loss': loss})
+
+        return loss, loss_dict
+
+    def p_mean_variance(self, x, c, t, clip_denoised: bool, return_codebook_ids=False, quantize_denoised=False,
+                        return_x0=False, score_corrector=None, corrector_kwargs=None):
+        t_in = t
+        model_out = self.apply_model(x, t_in, c, return_ids=return_codebook_ids)
+
+        if score_corrector is not None:
+            assert self.parameterization == "eps"
+            model_out = score_corrector.modify_score(self, model_out, x, t, c, **corrector_kwargs)
+
+        if return_codebook_ids:
+            model_out, logits = model_out
+
+        if self.parameterization == "eps":
+            x_recon = self.predict_start_from_noise(x, t=t, noise=model_out)
+        elif self.parameterization == "x0":
+            x_recon = model_out
+        else:
+            raise NotImplementedError()
+
+        if clip_denoised:
+            x_recon.clamp_(-1., 1.)
+        if quantize_denoised:
+            x_recon, _, [_, _, indices] = self.first_stage_model.quantize(x_recon)
+        model_mean, posterior_variance, posterior_log_variance = self.q_posterior(x_start=x_recon, x_t=x, t=t)
+        if return_codebook_ids:
+            return model_mean, posterior_variance, posterior_log_variance, logits
+        elif return_x0:
+            return model_mean, posterior_variance, posterior_log_variance, x_recon
+        else:
+            return model_mean, posterior_variance, posterior_log_variance
+
+    @torch.no_grad()
+    def p_sample(self, x, c, t, clip_denoised=False, repeat_noise=False,
+                 return_codebook_ids=False, quantize_denoised=False, return_x0=False,
+                 temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None):
+        b, *_, device = *x.shape, x.device
+        outputs = self.p_mean_variance(x=x, c=c, t=t, clip_denoised=clip_denoised,
+                                       return_codebook_ids=return_codebook_ids,
+                                       quantize_denoised=quantize_denoised,
+                                       return_x0=return_x0,
+                                       score_corrector=score_corrector, corrector_kwargs=corrector_kwargs)
+        if return_codebook_ids:
+            raise DeprecationWarning("Support dropped.")
+            model_mean, _, model_log_variance, logits = outputs
+        elif return_x0:
+            model_mean, _, model_log_variance, x0 = outputs
+        else:
+            model_mean, _, model_log_variance = outputs
+
+        noise = noise_like(x.shape, device, repeat_noise) * temperature
+        if noise_dropout > 0.:
+            noise = torch.nn.functional.dropout(noise, p=noise_dropout)
+        # no noise when t == 0
+        nonzero_mask = (1 - (t == 0).float()).reshape(b, *((1,) * (len(x.shape) - 1)))
+
+        if return_codebook_ids:
+            return model_mean + nonzero_mask * (0.5 * model_log_variance).exp() * noise, logits.argmax(dim=1)
+        if return_x0:
+            return model_mean + nonzero_mask * (0.5 * model_log_variance).exp() * noise, x0
+        else:
+            return model_mean + nonzero_mask * (0.5 * model_log_variance).exp() * noise
+
+    @torch.no_grad()
+    def progressive_denoising(self, cond, shape, verbose=True, callback=None, quantize_denoised=False,
+                              img_callback=None, mask=None, x0=None, temperature=1., noise_dropout=0.,
+                              score_corrector=None, corrector_kwargs=None, batch_size=None, x_T=None, start_T=None,
+                              log_every_t=None):
+        if not log_every_t:
+            log_every_t = self.log_every_t
+        timesteps = self.num_timesteps
+        if batch_size is not None:
+            b = batch_size if batch_size is not None else shape[0]
+            shape = [batch_size] + list(shape)
+        else:
+            b = batch_size = shape[0]
+        if x_T is None:
+            img = torch.randn(shape, device=self.device)
+        else:
+            img = x_T
+        intermediates = []
+        if cond is not None:
+            if isinstance(cond, dict):
+                cond = {key: cond[key][:batch_size] if not isinstance(cond[key], list) else
+                list(map(lambda x: x[:batch_size], cond[key])) for key in cond}
+            else:
+                cond = [c[:batch_size] for c in cond] if isinstance(cond, list) else cond[:batch_size]
+
+        if start_T is not None:
+            timesteps = min(timesteps, start_T)
+        iterator = tqdm(reversed(range(0, timesteps)), desc='Progressive Generation',
+                        total=timesteps) if verbose else reversed(
+            range(0, timesteps))
+        if type(temperature) == float:
+            temperature = [temperature] * timesteps
+
+        for i in iterator:
+            ts = torch.full((b,), i, device=self.device, dtype=torch.long)
+            if self.shorten_cond_schedule:
+                assert self.model.conditioning_key != 'hybrid'
+                tc = self.cond_ids[ts].to(cond.device)
+                cond = self.q_sample(x_start=cond, t=tc, noise=torch.randn_like(cond))
+
+            img, x0_partial = self.p_sample(img, cond, ts,
+                                            clip_denoised=self.clip_denoised,
+                                            quantize_denoised=quantize_denoised, return_x0=True,
+                                            temperature=temperature[i], noise_dropout=noise_dropout,
+                                            score_corrector=score_corrector, corrector_kwargs=corrector_kwargs)
+            if mask is not None:
+                assert x0 is not None
+                img_orig = self.q_sample(x0, ts)
+                img = img_orig * mask + (1. - mask) * img
+
+            if i % log_every_t == 0 or i == timesteps - 1:
+                intermediates.append(x0_partial)
+            if callback: callback(i)
+            if img_callback: img_callback(img, i)
+        return img, intermediates
+
+    @torch.no_grad()
+    def p_sample_loop(self, cond, shape, return_intermediates=False,
+                      x_T=None, verbose=True, callback=None, timesteps=None, quantize_denoised=False,
+                      mask=None, x0=None, img_callback=None, start_T=None,
+                      log_every_t=None):
+
+        if not log_every_t:
+            log_every_t = self.log_every_t
+        device = self.betas.device
+        b = shape[0]
+        if x_T is None:
+            img = torch.randn(shape, device=device)
+        else:
+            img = x_T
+
+        intermediates = [img]
+        if timesteps is None:
+            timesteps = self.num_timesteps
+
+        if start_T is not None:
+            timesteps = min(timesteps, start_T)
+        iterator = tqdm(reversed(range(0, timesteps)), desc='Sampling t', total=timesteps) if verbose else reversed(
+            range(0, timesteps))
+
+        if mask is not None:
+            assert x0 is not None
+            assert x0.shape[2:3] == mask.shape[2:3]  # spatial size has to match
+
+        for i in iterator:
+            ts = torch.full((b,), i, device=device, dtype=torch.long)
+            if self.shorten_cond_schedule:
+                assert self.model.conditioning_key != 'hybrid'
+                tc = self.cond_ids[ts].to(cond.device)
+                cond = self.q_sample(x_start=cond, t=tc, noise=torch.randn_like(cond))
+
+            img = self.p_sample(img, cond, ts,
+                                clip_denoised=self.clip_denoised,
+                                quantize_denoised=quantize_denoised)
+            if mask is not None:
+                img_orig = self.q_sample(x0, ts)
+                img = img_orig * mask + (1. - mask) * img
+
+            if i % log_every_t == 0 or i == timesteps - 1:
+                intermediates.append(img)
+            if callback: callback(i)
+            if img_callback: img_callback(img, i)
+
+        if return_intermediates:
+            return img, intermediates
+        return img
+
+    @torch.no_grad()
+    def sample(self, cond, batch_size=16, return_intermediates=False, x_T=None,
+               verbose=True, timesteps=None, quantize_denoised=False,
+               mask=None, x0=None, shape=None, **kwargs):
+        if shape is None:
+            shape = (batch_size, self.channels, self.image_size, self.image_size)
+        if cond is not None:
+            if isinstance(cond, dict):
+                cond = {key: cond[key][:batch_size] if not isinstance(cond[key], list) else
+                list(map(lambda x: x[:batch_size], cond[key])) for key in cond}
+            else:
+                cond = [c[:batch_size] for c in cond] if isinstance(cond, list) else cond[:batch_size]
+        return self.p_sample_loop(cond,
+                                  shape,
+                                  return_intermediates=return_intermediates, x_T=x_T,
+                                  verbose=verbose, timesteps=timesteps, quantize_denoised=quantize_denoised,
+                                  mask=mask, x0=x0)
+
+    @torch.no_grad()
+    def sample_log(self, cond, batch_size, ddim, ddim_steps, **kwargs):
+        if ddim:
+            ddim_sampler = DDIMSampler(self)
+            shape = (self.channels, self.image_size, self.image_size)
+            samples, intermediates = ddim_sampler.sample(ddim_steps, batch_size,
+                                                         shape, cond, verbose=False, **kwargs)
+
+        else:
+            samples, intermediates = self.sample(cond=cond, batch_size=batch_size,
+                                                 return_intermediates=True, **kwargs)
+
+        return samples, intermediates
+
+    @torch.no_grad()
+    def get_unconditional_conditioning(self, batch_size, null_label=None):
+        if null_label is not None:
+            xc = null_label
+            if isinstance(xc, ListConfig):
+                xc = list(xc)
+            if isinstance(xc, dict) or isinstance(xc, list):
+                c = self.get_learned_conditioning(xc)
+            else:
+                if hasattr(xc, "to"):
+                    xc = xc.to(self.device)
+                c = self.get_learned_conditioning(xc)
+        else:
+            if self.cond_stage_key in ["class_label", "cls"]:
+                xc = self.cond_stage_model.get_unconditional_conditioning(batch_size, device=self.device)
+                return self.get_learned_conditioning(xc)
+            else:
+                raise NotImplementedError("todo")
+        if isinstance(c, list):  # in case the encoder gives us a list
+            for i in range(len(c)):
+                c[i] = repeat(c[i], '1 ... -> b ...', b=batch_size).to(self.device)
+        else:
+            c = repeat(c, '1 ... -> b ...', b=batch_size).to(self.device)
+        return c
+
+    @torch.no_grad()
+    def log_images(self, batch, N=8, n_row=4, sample=True, ddim_steps=50, ddim_eta=0., return_keys=None,
+                   quantize_denoised=True, inpaint=True, plot_denoise_rows=False, plot_progressive_rows=True,
+                   plot_diffusion_rows=True, unconditional_guidance_scale=1., unconditional_guidance_label=None,
+                   use_ema_scope=True,
+                   **kwargs):
+        ema_scope = self.ema_scope if use_ema_scope else nullcontext
+        use_ddim = ddim_steps is not None
+
+        log = dict()
+        z, c, x, xrec, xc = self.get_input(batch, self.first_stage_key,
+                                           return_first_stage_outputs=True,
+                                           force_c_encode=True,
+                                           return_original_cond=True,
+                                           bs=N)
+        N = min(x.shape[0], N)
+        n_row = min(x.shape[0], n_row)
+        log["inputs"] = x
+        log["reconstruction"] = xrec
+        if self.model.conditioning_key is not None:
+            if hasattr(self.cond_stage_model, "decode"):
+                xc = self.cond_stage_model.decode(c)
+                log["conditioning"] = xc
+            elif self.cond_stage_key in ["caption", "txt"]:
+                xc = log_txt_as_img((x.shape[2], x.shape[3]), batch[self.cond_stage_key], size=x.shape[2] // 25)
+                log["conditioning"] = xc
+            elif self.cond_stage_key in ['class_label', "cls"]:
+                try:
+                    xc = log_txt_as_img((x.shape[2], x.shape[3]), batch["human_label"], size=x.shape[2] // 25)
+                    log['conditioning'] = xc
+                except KeyError:
+                    # probably no "human_label" in batch
+                    pass
+            elif isimage(xc):
+                log["conditioning"] = xc
+            if ismap(xc):
+                log["original_conditioning"] = self.to_rgb(xc)
+
+        if plot_diffusion_rows:
+            # get diffusion row
+            diffusion_row = list()
+            z_start = z[:n_row]
+            for t in range(self.num_timesteps):
+                if t % self.log_every_t == 0 or t == self.num_timesteps - 1:
+                    t = repeat(torch.tensor([t]), '1 -> b', b=n_row)
+                    t = t.to(self.device).long()
+                    noise = torch.randn_like(z_start)
+                    z_noisy = self.q_sample(x_start=z_start, t=t, noise=noise)
+                    diffusion_row.append(self.decode_first_stage(z_noisy))
+
+            diffusion_row = torch.stack(diffusion_row)  # n_log_step, n_row, C, H, W
+            diffusion_grid = rearrange(diffusion_row, 'n b c h w -> b n c h w')
+            diffusion_grid = rearrange(diffusion_grid, 'b n c h w -> (b n) c h w')
+            diffusion_grid = make_grid(diffusion_grid, nrow=diffusion_row.shape[0])
+            log["diffusion_row"] = diffusion_grid
+
+        if sample:
+            # get denoise row
+            with ema_scope("Sampling"):
+                samples, z_denoise_row = self.sample_log(cond=c, batch_size=N, ddim=use_ddim,
+                                                         ddim_steps=ddim_steps, eta=ddim_eta)
+                # samples, z_denoise_row = self.sample(cond=c, batch_size=N, return_intermediates=True)
+            x_samples = self.decode_first_stage(samples)
+            log["samples"] = x_samples
+            if plot_denoise_rows:
+                denoise_grid = self._get_denoise_row_from_list(z_denoise_row)
+                log["denoise_row"] = denoise_grid
+
+            if quantize_denoised and not isinstance(self.first_stage_model, AutoencoderKL) and not isinstance(
+                    self.first_stage_model, IdentityFirstStage):
+                # also display when quantizing x0 while sampling
+                with ema_scope("Plotting Quantized Denoised"):
+                    samples, z_denoise_row = self.sample_log(cond=c, batch_size=N, ddim=use_ddim,
+                                                             ddim_steps=ddim_steps, eta=ddim_eta,
+                                                             quantize_denoised=True)
+                    # samples, z_denoise_row = self.sample(cond=c, batch_size=N, return_intermediates=True,
+                    #                                      quantize_denoised=True)
+                x_samples = self.decode_first_stage(samples.to(self.device))
+                log["samples_x0_quantized"] = x_samples
+
+        if unconditional_guidance_scale > 1.0:
+            uc = self.get_unconditional_conditioning(N, unconditional_guidance_label)
+            if self.model.conditioning_key == "crossattn-adm":
+                uc = {"c_crossattn": [uc], "c_adm": c["c_adm"]}
+            with ema_scope("Sampling with classifier-free guidance"):
+                samples_cfg, _ = self.sample_log(cond=c, batch_size=N, ddim=use_ddim,
+                                                 ddim_steps=ddim_steps, eta=ddim_eta,
+                                                 unconditional_guidance_scale=unconditional_guidance_scale,
+                                                 unconditional_conditioning=uc,
+                                                 )
+                x_samples_cfg = self.decode_first_stage(samples_cfg)
+                log[f"samples_cfg_scale_{unconditional_guidance_scale:.2f}"] = x_samples_cfg
+
+        if inpaint:
+            # make a simple center square
+            b, h, w = z.shape[0], z.shape[2], z.shape[3]
+            mask = torch.ones(N, h, w).to(self.device)
+            # zeros will be filled in
+            mask[:, h // 4:3 * h // 4, w // 4:3 * w // 4] = 0.
+            mask = mask[:, None, ...]
+            with ema_scope("Plotting Inpaint"):
+                samples, _ = self.sample_log(cond=c, batch_size=N, ddim=use_ddim, eta=ddim_eta,
+                                             ddim_steps=ddim_steps, x0=z[:N], mask=mask)
+            x_samples = self.decode_first_stage(samples.to(self.device))
+            log["samples_inpainting"] = x_samples
+            log["mask"] = mask
+
+            # outpaint
+            mask = 1. - mask
+            with ema_scope("Plotting Outpaint"):
+                samples, _ = self.sample_log(cond=c, batch_size=N, ddim=use_ddim, eta=ddim_eta,
+                                             ddim_steps=ddim_steps, x0=z[:N], mask=mask)
+            x_samples = self.decode_first_stage(samples.to(self.device))
+            log["samples_outpainting"] = x_samples
+
+        if plot_progressive_rows:
+            with ema_scope("Plotting Progressives"):
+                img, progressives = self.progressive_denoising(c,
+                                                               shape=(self.channels, self.image_size, self.image_size),
+                                                               batch_size=N)
+            prog_row = self._get_denoise_row_from_list(progressives, desc="Progressive Generation")
+            log["progressive_row"] = prog_row
+
+        if return_keys:
+            if np.intersect1d(list(log.keys()), return_keys).shape[0] == 0:
+                return log
+            else:
+                return {key: log[key] for key in return_keys}
+        return log
+
+    def configure_optimizers(self):
+        lr = self.learning_rate
+        params = list(self.model.parameters())
+        if self.cond_stage_trainable:
+            print(f"{self.__class__.__name__}: Also optimizing conditioner params!")
+            params = params + list(self.cond_stage_model.parameters())
+        if self.learn_logvar:
+            print('Diffusion model optimizing logvar')
+            params.append(self.logvar)
+        opt = torch.optim.AdamW(params, lr=lr)
+        if self.use_scheduler:
+            assert 'target' in self.scheduler_config
+            scheduler = instantiate_from_config(self.scheduler_config)
+
+            print("Setting up LambdaLR scheduler...")
+            scheduler = [
+                {
+                    'scheduler': LambdaLR(opt, lr_lambda=scheduler.schedule),
+                    'interval': 'step',
+                    'frequency': 1
+                }]
+            return [opt], scheduler
+        return opt
+
+    @torch.no_grad()
+    def to_rgb(self, x):
+        x = x.float()
+        if not hasattr(self, "colorize"):
+            self.colorize = torch.randn(3, x.shape[1], 1, 1).to(x)
+        x = nn.functional.conv2d(x, weight=self.colorize)
+        x = 2. * (x - x.min()) / (x.max() - x.min()) - 1.
+        return x
+
+
+class DiffusionWrapper(pl.LightningModule):
+    def __init__(self, diff_model_config, conditioning_key):
+        super().__init__()
+        self.sequential_cross_attn = diff_model_config.pop("sequential_crossattn", False)
+        self.diffusion_model = instantiate_from_config(diff_model_config)
+        self.conditioning_key = conditioning_key
+        assert self.conditioning_key in [None, 'concat', 'crossattn', 'hybrid', 'adm', 'hybrid-adm', 'crossattn-adm']
+
+    def forward(self, x, t, c_concat: list = None, c_crossattn: list = None, c_adm=None):
+        if self.conditioning_key is None:
+            out = self.diffusion_model(x, t)
+        elif self.conditioning_key == 'concat':
+            xc = torch.cat([x] + c_concat, dim=1)
+            out = self.diffusion_model(xc, t)
+        elif self.conditioning_key == 'crossattn':
+            if not self.sequential_cross_attn:
+                cc = torch.cat(c_crossattn, 1)
+            else:
+                cc = c_crossattn
+            out = self.diffusion_model(x, t, context=cc)
+        elif self.conditioning_key == 'hybrid':
+            xc = torch.cat([x] + c_concat, dim=1)
+            cc = torch.cat(c_crossattn, 1)
+            out = self.diffusion_model(xc, t, context=cc)
+        elif self.conditioning_key == 'hybrid-adm':
+            assert c_adm is not None
+            xc = torch.cat([x] + c_concat, dim=1)
+            cc = torch.cat(c_crossattn, 1)
+            out = self.diffusion_model(xc, t, context=cc, y=c_adm)
+        elif self.conditioning_key == 'crossattn-adm':
+            assert c_adm is not None
+            cc = torch.cat(c_crossattn, 1)
+            out = self.diffusion_model(x, t, context=cc, y=c_adm)
+        elif self.conditioning_key == 'adm':
+            cc = c_crossattn[0]
+            out = self.diffusion_model(x, t, y=cc)
+        else:
+            raise NotImplementedError()
+
+        return out
+
+
+class LatentUpscaleDiffusion(LatentDiffusion):
+    def __init__(self, *args, low_scale_config, low_scale_key="LR", noise_level_key=None, **kwargs):
+        super().__init__(*args, **kwargs)
+        # assumes that neither the cond_stage nor the low_scale_model contain trainable params
+        assert not self.cond_stage_trainable
+        self.instantiate_low_stage(low_scale_config)
+        self.low_scale_key = low_scale_key
+        self.noise_level_key = noise_level_key
+
+    def instantiate_low_stage(self, config):
+        model = instantiate_from_config(config)
+        self.low_scale_model = model.eval()
+        self.low_scale_model.train = disabled_train
+        for param in self.low_scale_model.parameters():
+            param.requires_grad = False
+
+    @torch.no_grad()
+    def get_input(self, batch, k, cond_key=None, bs=None, log_mode=False):
+        if not log_mode:
+            z, c = super().get_input(batch, k, force_c_encode=True, bs=bs)
+        else:
+            z, c, x, xrec, xc = super().get_input(batch, self.first_stage_key, return_first_stage_outputs=True,
+                                                  force_c_encode=True, return_original_cond=True, bs=bs)
+        x_low = batch[self.low_scale_key][:bs]
+        x_low = rearrange(x_low, 'b h w c -> b c h w')
+        x_low = x_low.to(memory_format=torch.contiguous_format).float()
+        zx, noise_level = self.low_scale_model(x_low)
+        if self.noise_level_key is not None:
+            # get noise level from batch instead, e.g. when extracting a custom noise level for bsr
+            raise NotImplementedError('TODO')
+
+        all_conds = {"c_concat": [zx], "c_crossattn": [c], "c_adm": noise_level}
+        if log_mode:
+            # TODO: maybe disable if too expensive
+            x_low_rec = self.low_scale_model.decode(zx)
+            return z, all_conds, x, xrec, xc, x_low, x_low_rec, noise_level
+        return z, all_conds
+
+    @torch.no_grad()
+    def log_images(self, batch, N=8, n_row=4, sample=True, ddim_steps=200, ddim_eta=1., return_keys=None,
+                   plot_denoise_rows=False, plot_progressive_rows=True, plot_diffusion_rows=True,
+                   unconditional_guidance_scale=1., unconditional_guidance_label=None, use_ema_scope=True,
+                   **kwargs):
+        ema_scope = self.ema_scope if use_ema_scope else nullcontext
+        use_ddim = ddim_steps is not None
+
+        log = dict()
+        z, c, x, xrec, xc, x_low, x_low_rec, noise_level = self.get_input(batch, self.first_stage_key, bs=N,
+                                                                          log_mode=True)
+        N = min(x.shape[0], N)
+        n_row = min(x.shape[0], n_row)
+        log["inputs"] = x
+        log["reconstruction"] = xrec
+        log["x_lr"] = x_low
+        log[f"x_lr_rec_@noise_levels{'-'.join(map(lambda x: str(x), list(noise_level.cpu().numpy())))}"] = x_low_rec
+        if self.model.conditioning_key is not None:
+            if hasattr(self.cond_stage_model, "decode"):
+                xc = self.cond_stage_model.decode(c)
+                log["conditioning"] = xc
+            elif self.cond_stage_key in ["caption", "txt"]:
+                xc = log_txt_as_img((x.shape[2], x.shape[3]), batch[self.cond_stage_key], size=x.shape[2] // 25)
+                log["conditioning"] = xc
+            elif self.cond_stage_key in ['class_label', 'cls']:
+                xc = log_txt_as_img((x.shape[2], x.shape[3]), batch["human_label"], size=x.shape[2] // 25)
+                log['conditioning'] = xc
+            elif isimage(xc):
+                log["conditioning"] = xc
+            if ismap(xc):
+                log["original_conditioning"] = self.to_rgb(xc)
+
+        if plot_diffusion_rows:
+            # get diffusion row
+            diffusion_row = list()
+            z_start = z[:n_row]
+            for t in range(self.num_timesteps):
+                if t % self.log_every_t == 0 or t == self.num_timesteps - 1:
+                    t = repeat(torch.tensor([t]), '1 -> b', b=n_row)
+                    t = t.to(self.device).long()
+                    noise = torch.randn_like(z_start)
+                    z_noisy = self.q_sample(x_start=z_start, t=t, noise=noise)
+                    diffusion_row.append(self.decode_first_stage(z_noisy))
+
+            diffusion_row = torch.stack(diffusion_row)  # n_log_step, n_row, C, H, W
+            diffusion_grid = rearrange(diffusion_row, 'n b c h w -> b n c h w')
+            diffusion_grid = rearrange(diffusion_grid, 'b n c h w -> (b n) c h w')
+            diffusion_grid = make_grid(diffusion_grid, nrow=diffusion_row.shape[0])
+            log["diffusion_row"] = diffusion_grid
+
+        if sample:
+            # get denoise row
+            with ema_scope("Sampling"):
+                samples, z_denoise_row = self.sample_log(cond=c, batch_size=N, ddim=use_ddim,
+                                                         ddim_steps=ddim_steps, eta=ddim_eta)
+                # samples, z_denoise_row = self.sample(cond=c, batch_size=N, return_intermediates=True)
+            x_samples = self.decode_first_stage(samples)
+            log["samples"] = x_samples
+            if plot_denoise_rows:
+                denoise_grid = self._get_denoise_row_from_list(z_denoise_row)
+                log["denoise_row"] = denoise_grid
+
+        if unconditional_guidance_scale > 1.0:
+            uc_tmp = self.get_unconditional_conditioning(N, unconditional_guidance_label)
+            # TODO explore better "unconditional" choices for the other keys
+            # maybe guide away from empty text label and highest noise level and maximally degraded zx?
+            uc = dict()
+            for k in c:
+                if k == "c_crossattn":
+                    assert isinstance(c[k], list) and len(c[k]) == 1
+                    uc[k] = [uc_tmp]
+                elif k == "c_adm":  # todo: only run with text-based guidance?
+                    assert isinstance(c[k], torch.Tensor)
+                    #uc[k] = torch.ones_like(c[k]) * self.low_scale_model.max_noise_level
+                    uc[k] = c[k]
+                elif isinstance(c[k], list):
+                    uc[k] = [c[k][i] for i in range(len(c[k]))]
+                else:
+                    uc[k] = c[k]
+
+            with ema_scope("Sampling with classifier-free guidance"):
+                samples_cfg, _ = self.sample_log(cond=c, batch_size=N, ddim=use_ddim,
+                                                 ddim_steps=ddim_steps, eta=ddim_eta,
+                                                 unconditional_guidance_scale=unconditional_guidance_scale,
+                                                 unconditional_conditioning=uc,
+                                                 )
+                x_samples_cfg = self.decode_first_stage(samples_cfg)
+                log[f"samples_cfg_scale_{unconditional_guidance_scale:.2f}"] = x_samples_cfg
+
+        if plot_progressive_rows:
+            with ema_scope("Plotting Progressives"):
+                img, progressives = self.progressive_denoising(c,
+                                                               shape=(self.channels, self.image_size, self.image_size),
+                                                               batch_size=N)
+            prog_row = self._get_denoise_row_from_list(progressives, desc="Progressive Generation")
+            log["progressive_row"] = prog_row
+
+        return log
+
+
+class LatentFinetuneDiffusion(LatentDiffusion):
+    """
+         Basis for different finetunas, such as inpainting or depth2image
+         To disable finetuning mode, set finetune_keys to None
+    """
+
+    def __init__(self,
+                 concat_keys: tuple,
+                 finetune_keys=("model.diffusion_model.input_blocks.0.0.weight",
+                                "model_ema.diffusion_modelinput_blocks00weight"
+                                ),
+                 keep_finetune_dims=4,
+                 # if model was trained without concat mode before and we would like to keep these channels
+                 c_concat_log_start=None,  # to log reconstruction of c_concat codes
+                 c_concat_log_end=None,
+                 *args, **kwargs
+                 ):
+        ckpt_path = kwargs.pop("ckpt_path", None)
+        ignore_keys = kwargs.pop("ignore_keys", list())
+        super().__init__(*args, **kwargs)
+        self.finetune_keys = finetune_keys
+        self.concat_keys = concat_keys
+        self.keep_dims = keep_finetune_dims
+        self.c_concat_log_start = c_concat_log_start
+        self.c_concat_log_end = c_concat_log_end
+        if exists(self.finetune_keys): assert exists(ckpt_path), 'can only finetune from a given checkpoint'
+        if exists(ckpt_path):
+            self.init_from_ckpt(ckpt_path, ignore_keys)
+
+    def init_from_ckpt(self, path, ignore_keys=list(), only_model=False):
+        sd = torch.load(path, map_location="cpu")
+        if "state_dict" in list(sd.keys()):
+            sd = sd["state_dict"]
+        keys = list(sd.keys())
+        for k in keys:
+            for ik in ignore_keys:
+                if k.startswith(ik):
+                    print("Deleting key {} from state_dict.".format(k))
+                    del sd[k]
+
+            # make it explicit, finetune by including extra input channels
+            if exists(self.finetune_keys) and k in self.finetune_keys:
+                new_entry = None
+                for name, param in self.named_parameters():
+                    if name in self.finetune_keys:
+                        print(
+                            f"modifying key '{name}' and keeping its original {self.keep_dims} (channels) dimensions only")
+                        new_entry = torch.zeros_like(param)  # zero init
+                assert exists(new_entry), 'did not find matching parameter to modify'
+                new_entry[:, :self.keep_dims, ...] = sd[k]
+                sd[k] = new_entry
+
+        missing, unexpected = self.load_state_dict(sd, strict=False) if not only_model else self.model.load_state_dict(
+            sd, strict=False)
+        print(f"Restored from {path} with {len(missing)} missing and {len(unexpected)} unexpected keys")
+        if len(missing) > 0:
+            print(f"Missing Keys: {missing}")
+        if len(unexpected) > 0:
+            print(f"Unexpected Keys: {unexpected}")
+
+    @torch.no_grad()
+    def log_images(self, batch, N=8, n_row=4, sample=True, ddim_steps=200, ddim_eta=1., return_keys=None,
+                   quantize_denoised=True, inpaint=True, plot_denoise_rows=False, plot_progressive_rows=True,
+                   plot_diffusion_rows=True, unconditional_guidance_scale=1., unconditional_guidance_label=None,
+                   use_ema_scope=True,
+                   **kwargs):
+        ema_scope = self.ema_scope if use_ema_scope else nullcontext
+        use_ddim = ddim_steps is not None
+
+        log = dict()
+        z, c, x, xrec, xc = self.get_input(batch, self.first_stage_key, bs=N, return_first_stage_outputs=True)
+        c_cat, c = c["c_concat"][0], c["c_crossattn"][0]
+        N = min(x.shape[0], N)
+        n_row = min(x.shape[0], n_row)
+        log["inputs"] = x
+        log["reconstruction"] = xrec
+        if self.model.conditioning_key is not None:
+            if hasattr(self.cond_stage_model, "decode"):
+                xc = self.cond_stage_model.decode(c)
+                log["conditioning"] = xc
+            elif self.cond_stage_key in ["caption", "txt"]:
+                xc = log_txt_as_img((x.shape[2], x.shape[3]), batch[self.cond_stage_key], size=x.shape[2] // 25)
+                log["conditioning"] = xc
+            elif self.cond_stage_key in ['class_label', 'cls']:
+                xc = log_txt_as_img((x.shape[2], x.shape[3]), batch["human_label"], size=x.shape[2] // 25)
+                log['conditioning'] = xc
+            elif isimage(xc):
+                log["conditioning"] = xc
+            if ismap(xc):
+                log["original_conditioning"] = self.to_rgb(xc)
+
+        if not (self.c_concat_log_start is None and self.c_concat_log_end is None):
+            log["c_concat_decoded"] = self.decode_first_stage(c_cat[:, self.c_concat_log_start:self.c_concat_log_end])
+
+        if plot_diffusion_rows:
+            # get diffusion row
+            diffusion_row = list()
+            z_start = z[:n_row]
+            for t in range(self.num_timesteps):
+                if t % self.log_every_t == 0 or t == self.num_timesteps - 1:
+                    t = repeat(torch.tensor([t]), '1 -> b', b=n_row)
+                    t = t.to(self.device).long()
+                    noise = torch.randn_like(z_start)
+                    z_noisy = self.q_sample(x_start=z_start, t=t, noise=noise)
+                    diffusion_row.append(self.decode_first_stage(z_noisy))
+
+            diffusion_row = torch.stack(diffusion_row)  # n_log_step, n_row, C, H, W
+            diffusion_grid = rearrange(diffusion_row, 'n b c h w -> b n c h w')
+            diffusion_grid = rearrange(diffusion_grid, 'b n c h w -> (b n) c h w')
+            diffusion_grid = make_grid(diffusion_grid, nrow=diffusion_row.shape[0])
+            log["diffusion_row"] = diffusion_grid
+
+        if sample:
+            # get denoise row
+            with ema_scope("Sampling"):
+                samples, z_denoise_row = self.sample_log(cond={"c_concat": [c_cat], "c_crossattn": [c]},
+                                                         batch_size=N, ddim=use_ddim,
+                                                         ddim_steps=ddim_steps, eta=ddim_eta)
+                # samples, z_denoise_row = self.sample(cond=c, batch_size=N, return_intermediates=True)
+            x_samples = self.decode_first_stage(samples)
+            log["samples"] = x_samples
+            if plot_denoise_rows:
+                denoise_grid = self._get_denoise_row_from_list(z_denoise_row)
+                log["denoise_row"] = denoise_grid
+
+        if unconditional_guidance_scale > 1.0:
+            uc_cross = self.get_unconditional_conditioning(N, unconditional_guidance_label)
+            uc_cat = c_cat
+            uc_full = {"c_concat": [uc_cat], "c_crossattn": [uc_cross]}
+            with ema_scope("Sampling with classifier-free guidance"):
+                samples_cfg, _ = self.sample_log(cond={"c_concat": [c_cat], "c_crossattn": [c]},
+                                                 batch_size=N, ddim=use_ddim,
+                                                 ddim_steps=ddim_steps, eta=ddim_eta,
+                                                 unconditional_guidance_scale=unconditional_guidance_scale,
+                                                 unconditional_conditioning=uc_full,
+                                                 )
+                x_samples_cfg = self.decode_first_stage(samples_cfg)
+                log[f"samples_cfg_scale_{unconditional_guidance_scale:.2f}"] = x_samples_cfg
+
+        return log
+
+
+class LatentInpaintDiffusion(LatentFinetuneDiffusion):
+    """
+    can either run as pure inpainting model (only concat mode) or with mixed conditionings,
+    e.g. mask as concat and text via cross-attn.
+    To disable finetuning mode, set finetune_keys to None
+     """
+
+    def __init__(self,
+                 concat_keys=("mask", "masked_image"),
+                 masked_image_key="masked_image",
+                 *args, **kwargs
+                 ):
+        super().__init__(concat_keys, *args, **kwargs)
+        self.masked_image_key = masked_image_key
+        assert self.masked_image_key in concat_keys
+
+    @torch.no_grad()
+    def get_input(self, batch, k, cond_key=None, bs=None, return_first_stage_outputs=False):
+        # note: restricted to non-trainable encoders currently
+        assert not self.cond_stage_trainable, 'trainable cond stages not yet supported for inpainting'
+        z, c, x, xrec, xc = super().get_input(batch, self.first_stage_key, return_first_stage_outputs=True,
+                                              force_c_encode=True, return_original_cond=True, bs=bs)
+
+        assert exists(self.concat_keys)
+        c_cat = list()
+        for ck in self.concat_keys:
+            cc = rearrange(batch[ck], 'b h w c -> b c h w').to(memory_format=torch.contiguous_format).float()
+            if bs is not None:
+                cc = cc[:bs]
+                cc = cc.to(self.device)
+            bchw = z.shape
+            if ck != self.masked_image_key:
+                cc = torch.nn.functional.interpolate(cc, size=bchw[-2:])
+            else:
+                cc = self.get_first_stage_encoding(self.encode_first_stage(cc))
+            c_cat.append(cc)
+        c_cat = torch.cat(c_cat, dim=1)
+        all_conds = {"c_concat": [c_cat], "c_crossattn": [c]}
+        if return_first_stage_outputs:
+            return z, all_conds, x, xrec, xc
+        return z, all_conds
+
+    @torch.no_grad()
+    def log_images(self, *args, **kwargs):
+        log = super(LatentInpaintDiffusion, self).log_images(*args, **kwargs)
+        log["masked_image"] = rearrange(args[0]["masked_image"],
+                                        'b h w c -> b c h w').to(memory_format=torch.contiguous_format).float()
+        return log
+
+
+class LatentDepth2ImageDiffusion(LatentFinetuneDiffusion):
+    """
+    condition on monocular depth estimation
+    """
+
+    def __init__(self, depth_stage_config, concat_keys=("midas_in",), *args, **kwargs):
+        super().__init__(concat_keys=concat_keys, *args, **kwargs)
+        self.depth_model = instantiate_from_config(depth_stage_config)
+        self.depth_stage_key = concat_keys[0]
+
+    @torch.no_grad()
+    def get_input(self, batch, k, cond_key=None, bs=None, return_first_stage_outputs=False):
+        # note: restricted to non-trainable encoders currently
+        assert not self.cond_stage_trainable, 'trainable cond stages not yet supported for depth2img'
+        z, c, x, xrec, xc = super().get_input(batch, self.first_stage_key, return_first_stage_outputs=True,
+                                              force_c_encode=True, return_original_cond=True, bs=bs)
+
+        assert exists(self.concat_keys)
+        assert len(self.concat_keys) == 1
+        c_cat = list()
+        for ck in self.concat_keys:
+            cc = batch[ck]
+            if bs is not None:
+                cc = cc[:bs]
+                cc = cc.to(self.device)
+            cc = self.depth_model(cc)
+            cc = torch.nn.functional.interpolate(
+                cc,
+                size=z.shape[2:],
+                mode="bicubic",
+                align_corners=False,
+            )
+
+            depth_min, depth_max = torch.amin(cc, dim=[1, 2, 3], keepdim=True), torch.amax(cc, dim=[1, 2, 3],
+                                                                                           keepdim=True)
+            cc = 2. * (cc - depth_min) / (depth_max - depth_min + 0.001) - 1.
+            c_cat.append(cc)
+        c_cat = torch.cat(c_cat, dim=1)
+        all_conds = {"c_concat": [c_cat], "c_crossattn": [c]}
+        if return_first_stage_outputs:
+            return z, all_conds, x, xrec, xc
+        return z, all_conds
+
+    @torch.no_grad()
+    def log_images(self, *args, **kwargs):
+        log = super().log_images(*args, **kwargs)
+        depth = self.depth_model(args[0][self.depth_stage_key])
+        depth_min, depth_max = torch.amin(depth, dim=[1, 2, 3], keepdim=True), \
+                               torch.amax(depth, dim=[1, 2, 3], keepdim=True)
+        log["depth"] = 2. * (depth - depth_min) / (depth_max - depth_min) - 1.
+        return log
+
+
+class LatentUpscaleFinetuneDiffusion(LatentFinetuneDiffusion):
+    """
+        condition on low-res image (and optionally on some spatial noise augmentation)
+    """
+    def __init__(self, concat_keys=("lr",), reshuffle_patch_size=None,
+                 low_scale_config=None, low_scale_key=None, *args, **kwargs):
+        super().__init__(concat_keys=concat_keys, *args, **kwargs)
+        self.reshuffle_patch_size = reshuffle_patch_size
+        self.low_scale_model = None
+        if low_scale_config is not None:
+            print("Initializing a low-scale model")
+            assert exists(low_scale_key)
+            self.instantiate_low_stage(low_scale_config)
+            self.low_scale_key = low_scale_key
+
+    def instantiate_low_stage(self, config):
+        model = instantiate_from_config(config)
+        self.low_scale_model = model.eval()
+        self.low_scale_model.train = disabled_train
+        for param in self.low_scale_model.parameters():
+            param.requires_grad = False
+
+    @torch.no_grad()
+    def get_input(self, batch, k, cond_key=None, bs=None, return_first_stage_outputs=False):
+        # note: restricted to non-trainable encoders currently
+        assert not self.cond_stage_trainable, 'trainable cond stages not yet supported for upscaling-ft'
+        z, c, x, xrec, xc = super().get_input(batch, self.first_stage_key, return_first_stage_outputs=True,
+                                              force_c_encode=True, return_original_cond=True, bs=bs)
+
+        assert exists(self.concat_keys)
+        assert len(self.concat_keys) == 1
+        # optionally make spatial noise_level here
+        c_cat = list()
+        noise_level = None
+        for ck in self.concat_keys:
+            cc = batch[ck]
+            cc = rearrange(cc, 'b h w c -> b c h w')
+            if exists(self.reshuffle_patch_size):
+                assert isinstance(self.reshuffle_patch_size, int)
+                cc = rearrange(cc, 'b c (p1 h) (p2 w) -> b (p1 p2 c) h w',
+                               p1=self.reshuffle_patch_size, p2=self.reshuffle_patch_size)
+            if bs is not None:
+                cc = cc[:bs]
+                cc = cc.to(self.device)
+            if exists(self.low_scale_model) and ck == self.low_scale_key:
+                cc, noise_level = self.low_scale_model(cc)
+            c_cat.append(cc)
+        c_cat = torch.cat(c_cat, dim=1)
+        if exists(noise_level):
+            all_conds = {"c_concat": [c_cat], "c_crossattn": [c], "c_adm": noise_level}
+        else:
+            all_conds = {"c_concat": [c_cat], "c_crossattn": [c]}
+        if return_first_stage_outputs:
+            return z, all_conds, x, xrec, xc
+        return z, all_conds
+
+    @torch.no_grad()
+    def log_images(self, *args, **kwargs):
+        log = super().log_images(*args, **kwargs)
+        log["lr"] = rearrange(args[0]["lr"], 'b h w c -> b c h w')
+        return log

ldm/models/diffusion/dpm_solver/__init__.py

@@ -0,0 +1 @@
+from .sampler import DPMSolverSampler

ldm/models/diffusion/dpm_solver/dpm_solver.py

@@ -0,0 +1,1154 @@
+import torch
+import torch.nn.functional as F
+import math
+from tqdm import tqdm
+
+
+class NoiseScheduleVP:
+    def __init__(
+            self,
+            schedule='discrete',
+            betas=None,
+            alphas_cumprod=None,
+            continuous_beta_0=0.1,
+            continuous_beta_1=20.,
+    ):
+        """Create a wrapper class for the forward SDE (VP type).
+        ***
+        Update: We support discrete-time diffusion models by implementing a picewise linear interpolation for log_alpha_t.
+                We recommend to use schedule='discrete' for the discrete-time diffusion models, especially for high-resolution images.
+        ***
+        The forward SDE ensures that the condition distribution q_{t|0}(x_t | x_0) = N ( alpha_t * x_0, sigma_t^2 * I ).
+        We further define lambda_t = log(alpha_t) - log(sigma_t), which is the half-logSNR (described in the DPM-Solver paper).
+        Therefore, we implement the functions for computing alpha_t, sigma_t and lambda_t. For t in [0, T], we have:
+            log_alpha_t = self.marginal_log_mean_coeff(t)
+            sigma_t = self.marginal_std(t)
+            lambda_t = self.marginal_lambda(t)
+        Moreover, as lambda(t) is an invertible function, we also support its inverse function:
+            t = self.inverse_lambda(lambda_t)
+        ===============================================================
+        We support both discrete-time DPMs (trained on n = 0, 1, ..., N-1) and continuous-time DPMs (trained on t in [t_0, T]).
+        1. For discrete-time DPMs:
+            For discrete-time DPMs trained on n = 0, 1, ..., N-1, we convert the discrete steps to continuous time steps by:
+                t_i = (i + 1) / N
+            e.g. for N = 1000, we have t_0 = 1e-3 and T = t_{N-1} = 1.
+            We solve the corresponding diffusion ODE from time T = 1 to time t_0 = 1e-3.
+            Args:
+                betas: A `torch.Tensor`. The beta array for the discrete-time DPM. (See the original DDPM paper for details)
+                alphas_cumprod: A `torch.Tensor`. The cumprod alphas for the discrete-time DPM. (See the original DDPM paper for details)
+            Note that we always have alphas_cumprod = cumprod(betas). Therefore, we only need to set one of `betas` and `alphas_cumprod`.
+            **Important**:  Please pay special attention for the args for `alphas_cumprod`:
+                The `alphas_cumprod` is the \hat{alpha_n} arrays in the notations of DDPM. Specifically, DDPMs assume that
+                    q_{t_n | 0}(x_{t_n} | x_0) = N ( \sqrt{\hat{alpha_n}} * x_0, (1 - \hat{alpha_n}) * I ).
+                Therefore, the notation \hat{alpha_n} is different from the notation alpha_t in DPM-Solver. In fact, we have
+                    alpha_{t_n} = \sqrt{\hat{alpha_n}},
+                and
+                    log(alpha_{t_n}) = 0.5 * log(\hat{alpha_n}).
+        2. For continuous-time DPMs:
+            We support two types of VPSDEs: linear (DDPM) and cosine (improved-DDPM). The hyperparameters for the noise
+            schedule are the default settings in DDPM and improved-DDPM:
+            Args:
+                beta_min: A `float` number. The smallest beta for the linear schedule.
+                beta_max: A `float` number. The largest beta for the linear schedule.
+                cosine_s: A `float` number. The hyperparameter in the cosine schedule.
+                cosine_beta_max: A `float` number. The hyperparameter in the cosine schedule.
+                T: A `float` number. The ending time of the forward process.
+        ===============================================================
+        Args:
+            schedule: A `str`. The noise schedule of the forward SDE. 'discrete' for discrete-time DPMs,
+                    'linear' or 'cosine' for continuous-time DPMs.
+        Returns:
+            A wrapper object of the forward SDE (VP type).
+
+        ===============================================================
+        Example:
+        # For discrete-time DPMs, given betas (the beta array for n = 0, 1, ..., N - 1):
+        >>> ns = NoiseScheduleVP('discrete', betas=betas)
+        # For discrete-time DPMs, given alphas_cumprod (the \hat{alpha_n} array for n = 0, 1, ..., N - 1):
+        >>> ns = NoiseScheduleVP('discrete', alphas_cumprod=alphas_cumprod)
+        # For continuous-time DPMs (VPSDE), linear schedule:
+        >>> ns = NoiseScheduleVP('linear', continuous_beta_0=0.1, continuous_beta_1=20.)
+        """
+
+        if schedule not in ['discrete', 'linear', 'cosine']:
+            raise ValueError(
+                "Unsupported noise schedule {}. The schedule needs to be 'discrete' or 'linear' or 'cosine'".format(
+                    schedule))
+
+        self.schedule = schedule
+        if schedule == 'discrete':
+            if betas is not None:
+                log_alphas = 0.5 * torch.log(1 - betas).cumsum(dim=0)
+            else:
+                assert alphas_cumprod is not None
+                log_alphas = 0.5 * torch.log(alphas_cumprod)
+            self.total_N = len(log_alphas)
+            self.T = 1.
+            self.t_array = torch.linspace(0., 1., self.total_N + 1)[1:].reshape((1, -1))
+            self.log_alpha_array = log_alphas.reshape((1, -1,))
+        else:
+            self.total_N = 1000
+            self.beta_0 = continuous_beta_0
+            self.beta_1 = continuous_beta_1
+            self.cosine_s = 0.008
+            self.cosine_beta_max = 999.
+            self.cosine_t_max = math.atan(self.cosine_beta_max * (1. + self.cosine_s) / math.pi) * 2. * (
+                        1. + self.cosine_s) / math.pi - self.cosine_s
+            self.cosine_log_alpha_0 = math.log(math.cos(self.cosine_s / (1. + self.cosine_s) * math.pi / 2.))
+            self.schedule = schedule
+            if schedule == 'cosine':
+                # For the cosine schedule, T = 1 will have numerical issues. So we manually set the ending time T.
+                # Note that T = 0.9946 may be not the optimal setting. However, we find it works well.
+                self.T = 0.9946
+            else:
+                self.T = 1.
+
+    def marginal_log_mean_coeff(self, t):
+        """
+        Compute log(alpha_t) of a given continuous-time label t in [0, T].
+        """
+        if self.schedule == 'discrete':
+            return interpolate_fn(t.reshape((-1, 1)), self.t_array.to(t.device),
+                                  self.log_alpha_array.to(t.device)).reshape((-1))
+        elif self.schedule == 'linear':
+            return -0.25 * t ** 2 * (self.beta_1 - self.beta_0) - 0.5 * t * self.beta_0
+        elif self.schedule == 'cosine':
+            log_alpha_fn = lambda s: torch.log(torch.cos((s + self.cosine_s) / (1. + self.cosine_s) * math.pi / 2.))
+            log_alpha_t = log_alpha_fn(t) - self.cosine_log_alpha_0
+            return log_alpha_t
+
+    def marginal_alpha(self, t):
+        """
+        Compute alpha_t of a given continuous-time label t in [0, T].
+        """
+        return torch.exp(self.marginal_log_mean_coeff(t))
+
+    def marginal_std(self, t):
+        """
+        Compute sigma_t of a given continuous-time label t in [0, T].
+        """
+        return torch.sqrt(1. - torch.exp(2. * self.marginal_log_mean_coeff(t)))
+
+    def marginal_lambda(self, t):
+        """
+        Compute lambda_t = log(alpha_t) - log(sigma_t) of a given continuous-time label t in [0, T].
+        """
+        log_mean_coeff = self.marginal_log_mean_coeff(t)
+        log_std = 0.5 * torch.log(1. - torch.exp(2. * log_mean_coeff))
+        return log_mean_coeff - log_std
+
+    def inverse_lambda(self, lamb):
+        """
+        Compute the continuous-time label t in [0, T] of a given half-logSNR lambda_t.
+        """
+        if self.schedule == 'linear':
+            tmp = 2. * (self.beta_1 - self.beta_0) * torch.logaddexp(-2. * lamb, torch.zeros((1,)).to(lamb))
+            Delta = self.beta_0 ** 2 + tmp
+            return tmp / (torch.sqrt(Delta) + self.beta_0) / (self.beta_1 - self.beta_0)
+        elif self.schedule == 'discrete':
+            log_alpha = -0.5 * torch.logaddexp(torch.zeros((1,)).to(lamb.device), -2. * lamb)
+            t = interpolate_fn(log_alpha.reshape((-1, 1)), torch.flip(self.log_alpha_array.to(lamb.device), [1]),
+                               torch.flip(self.t_array.to(lamb.device), [1]))
+            return t.reshape((-1,))
+        else:
+            log_alpha = -0.5 * torch.logaddexp(-2. * lamb, torch.zeros((1,)).to(lamb))
+            t_fn = lambda log_alpha_t: torch.arccos(torch.exp(log_alpha_t + self.cosine_log_alpha_0)) * 2. * (
+                        1. + self.cosine_s) / math.pi - self.cosine_s
+            t = t_fn(log_alpha)
+            return t
+
+
+def model_wrapper(
+        model,
+        noise_schedule,
+        model_type="noise",
+        model_kwargs={},
+        guidance_type="uncond",
+        condition=None,
+        unconditional_condition=None,
+        guidance_scale=1.,
+        classifier_fn=None,
+        classifier_kwargs={},
+):
+    """Create a wrapper function for the noise prediction model.
+    DPM-Solver needs to solve the continuous-time diffusion ODEs. For DPMs trained on discrete-time labels, we need to
+    firstly wrap the model function to a noise prediction model that accepts the continuous time as the input.
+    We support four types of the diffusion model by setting `model_type`:
+        1. "noise": noise prediction model. (Trained by predicting noise).
+        2. "x_start": data prediction model. (Trained by predicting the data x_0 at time 0).
+        3. "v": velocity prediction model. (Trained by predicting the velocity).
+            The "v" prediction is derivation detailed in Appendix D of [1], and is used in Imagen-Video [2].
+            [1] Salimans, Tim, and Jonathan Ho. "Progressive distillation for fast sampling of diffusion models."
+                arXiv preprint arXiv:2202.00512 (2022).
+            [2] Ho, Jonathan, et al. "Imagen Video: High Definition Video Generation with Diffusion Models."
+                arXiv preprint arXiv:2210.02303 (2022).
+
+        4. "score": marginal score function. (Trained by denoising score matching).
+            Note that the score function and the noise prediction model follows a simple relationship:
+            ```
+                noise(x_t, t) = -sigma_t * score(x_t, t)
+            ```
+    We support three types of guided sampling by DPMs by setting `guidance_type`:
+        1. "uncond": unconditional sampling by DPMs.
+            The input `model` has the following format:
+            ``
+                model(x, t_input, **model_kwargs) -> noise | x_start | v | score
+            ``
+        2. "classifier": classifier guidance sampling [3] by DPMs and another classifier.
+            The input `model` has the following format:
+            ``
+                model(x, t_input, **model_kwargs) -> noise | x_start | v | score
+            ``
+            The input `classifier_fn` has the following format:
+            ``
+                classifier_fn(x, t_input, cond, **classifier_kwargs) -> logits(x, t_input, cond)
+            ``
+            [3] P. Dhariwal and A. Q. Nichol, "Diffusion models beat GANs on image synthesis,"
+                in Advances in Neural Information Processing Systems, vol. 34, 2021, pp. 8780-8794.
+        3. "classifier-free": classifier-free guidance sampling by conditional DPMs.
+            The input `model` has the following format:
+            ``
+                model(x, t_input, cond, **model_kwargs) -> noise | x_start | v | score
+            ``
+            And if cond == `unconditional_condition`, the model output is the unconditional DPM output.
+            [4] Ho, Jonathan, and Tim Salimans. "Classifier-free diffusion guidance."
+                arXiv preprint arXiv:2207.12598 (2022).
+
+    The `t_input` is the time label of the model, which may be discrete-time labels (i.e. 0 to 999)
+    or continuous-time labels (i.e. epsilon to T).
+    We wrap the model function to accept only `x` and `t_continuous` as inputs, and outputs the predicted noise:
+    ``
+        def model_fn(x, t_continuous) -> noise:
+            t_input = get_model_input_time(t_continuous)
+            return noise_pred(model, x, t_input, **model_kwargs)
+    ``
+    where `t_continuous` is the continuous time labels (i.e. epsilon to T). And we use `model_fn` for DPM-Solver.
+    ===============================================================
+    Args:
+        model: A diffusion model with the corresponding format described above.
+        noise_schedule: A noise schedule object, such as NoiseScheduleVP.
+        model_type: A `str`. The parameterization type of the diffusion model.
+                    "noise" or "x_start" or "v" or "score".
+        model_kwargs: A `dict`. A dict for the other inputs of the model function.
+        guidance_type: A `str`. The type of the guidance for sampling.
+                    "uncond" or "classifier" or "classifier-free".
+        condition: A pytorch tensor. The condition for the guided sampling.
+                    Only used for "classifier" or "classifier-free" guidance type.
+        unconditional_condition: A pytorch tensor. The condition for the unconditional sampling.
+                    Only used for "classifier-free" guidance type.
+        guidance_scale: A `float`. The scale for the guided sampling.
+        classifier_fn: A classifier function. Only used for the classifier guidance.
+        classifier_kwargs: A `dict`. A dict for the other inputs of the classifier function.
+    Returns:
+        A noise prediction model that accepts the noised data and the continuous time as the inputs.
+    """
+
+    def get_model_input_time(t_continuous):
+        """
+        Convert the continuous-time `t_continuous` (in [epsilon, T]) to the model input time.
+        For discrete-time DPMs, we convert `t_continuous` in [1 / N, 1] to `t_input` in [0, 1000 * (N - 1) / N].
+        For continuous-time DPMs, we just use `t_continuous`.
+        """
+        if noise_schedule.schedule == 'discrete':
+            return (t_continuous - 1. / noise_schedule.total_N) * 1000.
+        else:
+            return t_continuous
+
+    def noise_pred_fn(x, t_continuous, cond=None):
+        if t_continuous.reshape((-1,)).shape[0] == 1:
+            t_continuous = t_continuous.expand((x.shape[0]))
+        t_input = get_model_input_time(t_continuous)
+        if cond is None:
+            output = model(x, t_input, **model_kwargs)
+        else:
+            output = model(x, t_input, cond, **model_kwargs)
+        if model_type == "noise":
+            return output
+        elif model_type == "x_start":
+            alpha_t, sigma_t = noise_schedule.marginal_alpha(t_continuous), noise_schedule.marginal_std(t_continuous)
+            dims = x.dim()
+            return (x - expand_dims(alpha_t, dims) * output) / expand_dims(sigma_t, dims)
+        elif model_type == "v":
+            alpha_t, sigma_t = noise_schedule.marginal_alpha(t_continuous), noise_schedule.marginal_std(t_continuous)
+            dims = x.dim()
+            return expand_dims(alpha_t, dims) * output + expand_dims(sigma_t, dims) * x
+        elif model_type == "score":
+            sigma_t = noise_schedule.marginal_std(t_continuous)
+            dims = x.dim()
+            return -expand_dims(sigma_t, dims) * output
+
+    def cond_grad_fn(x, t_input):
+        """
+        Compute the gradient of the classifier, i.e. nabla_{x} log p_t(cond | x_t).
+        """
+        with torch.enable_grad():
+            x_in = x.detach().requires_grad_(True)
+            log_prob = classifier_fn(x_in, t_input, condition, **classifier_kwargs)
+            return torch.autograd.grad(log_prob.sum(), x_in)[0]
+
+    def model_fn(x, t_continuous):
+        """
+        The noise predicition model function that is used for DPM-Solver.
+        """
+        if t_continuous.reshape((-1,)).shape[0] == 1:
+            t_continuous = t_continuous.expand((x.shape[0]))
+        if guidance_type == "uncond":
+            return noise_pred_fn(x, t_continuous)
+        elif guidance_type == "classifier":
+            assert classifier_fn is not None
+            t_input = get_model_input_time(t_continuous)
+            cond_grad = cond_grad_fn(x, t_input)
+            sigma_t = noise_schedule.marginal_std(t_continuous)
+            noise = noise_pred_fn(x, t_continuous)
+            return noise - guidance_scale * expand_dims(sigma_t, dims=cond_grad.dim()) * cond_grad
+        elif guidance_type == "classifier-free":
+            if guidance_scale == 1. or unconditional_condition is None:
+                return noise_pred_fn(x, t_continuous, cond=condition)
+            else:
+                x_in = torch.cat([x] * 2)
+                t_in = torch.cat([t_continuous] * 2)
+                c_in = torch.cat([unconditional_condition, condition])
+                noise_uncond, noise = noise_pred_fn(x_in, t_in, cond=c_in).chunk(2)
+                return noise_uncond + guidance_scale * (noise - noise_uncond)
+
+    assert model_type in ["noise", "x_start", "v"]
+    assert guidance_type in ["uncond", "classifier", "classifier-free"]
+    return model_fn
+
+
+class DPM_Solver:
+    def __init__(self, model_fn, noise_schedule, predict_x0=False, thresholding=False, max_val=1.):
+        """Construct a DPM-Solver.
+        We support both the noise prediction model ("predicting epsilon") and the data prediction model ("predicting x0").
+        If `predict_x0` is False, we use the solver for the noise prediction model (DPM-Solver).
+        If `predict_x0` is True, we use the solver for the data prediction model (DPM-Solver++).
+            In such case, we further support the "dynamic thresholding" in [1] when `thresholding` is True.
+            The "dynamic thresholding" can greatly improve the sample quality for pixel-space DPMs with large guidance scales.
+        Args:
+            model_fn: A noise prediction model function which accepts the continuous-time input (t in [epsilon, T]):
+                ``
+                def model_fn(x, t_continuous):
+                    return noise
+                ``
+            noise_schedule: A noise schedule object, such as NoiseScheduleVP.
+            predict_x0: A `bool`. If true, use the data prediction model; else, use the noise prediction model.
+            thresholding: A `bool`. Valid when `predict_x0` is True. Whether to use the "dynamic thresholding" in [1].
+            max_val: A `float`. Valid when both `predict_x0` and `thresholding` are True. The max value for thresholding.
+
+        [1] Chitwan Saharia, William Chan, Saurabh Saxena, Lala Li, Jay Whang, Emily Denton, Seyed Kamyar Seyed Ghasemipour, Burcu Karagol Ayan, S Sara Mahdavi, Rapha Gontijo Lopes, et al. Photorealistic text-to-image diffusion models with deep language understanding. arXiv preprint arXiv:2205.11487, 2022b.
+        """
+        self.model = model_fn
+        self.noise_schedule = noise_schedule
+        self.predict_x0 = predict_x0
+        self.thresholding = thresholding
+        self.max_val = max_val
+
+    def noise_prediction_fn(self, x, t):
+        """
+        Return the noise prediction model.
+        """
+        return self.model(x, t)
+
+    def data_prediction_fn(self, x, t):
+        """
+        Return the data prediction model (with thresholding).
+        """
+        noise = self.noise_prediction_fn(x, t)
+        dims = x.dim()
+        alpha_t, sigma_t = self.noise_schedule.marginal_alpha(t), self.noise_schedule.marginal_std(t)
+        x0 = (x - expand_dims(sigma_t, dims) * noise) / expand_dims(alpha_t, dims)
+        if self.thresholding:
+            p = 0.995  # A hyperparameter in the paper of "Imagen" [1].
+            s = torch.quantile(torch.abs(x0).reshape((x0.shape[0], -1)), p, dim=1)
+            s = expand_dims(torch.maximum(s, self.max_val * torch.ones_like(s).to(s.device)), dims)
+            x0 = torch.clamp(x0, -s, s) / s
+        return x0
+
+    def model_fn(self, x, t):
+        """
+        Convert the model to the noise prediction model or the data prediction model.
+        """
+        if self.predict_x0:
+            return self.data_prediction_fn(x, t)
+        else:
+            return self.noise_prediction_fn(x, t)
+
+    def get_time_steps(self, skip_type, t_T, t_0, N, device):
+        """Compute the intermediate time steps for sampling.
+        Args:
+            skip_type: A `str`. The type for the spacing of the time steps. We support three types:
+                - 'logSNR': uniform logSNR for the time steps.
+                - 'time_uniform': uniform time for the time steps. (**Recommended for high-resolutional data**.)
+                - 'time_quadratic': quadratic time for the time steps. (Used in DDIM for low-resolutional data.)
+            t_T: A `float`. The starting time of the sampling (default is T).
+            t_0: A `float`. The ending time of the sampling (default is epsilon).
+            N: A `int`. The total number of the spacing of the time steps.
+            device: A torch device.
+        Returns:
+            A pytorch tensor of the time steps, with the shape (N + 1,).
+        """
+        if skip_type == 'logSNR':
+            lambda_T = self.noise_schedule.marginal_lambda(torch.tensor(t_T).to(device))
+            lambda_0 = self.noise_schedule.marginal_lambda(torch.tensor(t_0).to(device))
+            logSNR_steps = torch.linspace(lambda_T.cpu().item(), lambda_0.cpu().item(), N + 1).to(device)
+            return self.noise_schedule.inverse_lambda(logSNR_steps)
+        elif skip_type == 'time_uniform':
+            return torch.linspace(t_T, t_0, N + 1).to(device)
+        elif skip_type == 'time_quadratic':
+            t_order = 2
+            t = torch.linspace(t_T ** (1. / t_order), t_0 ** (1. / t_order), N + 1).pow(t_order).to(device)
+            return t
+        else:
+            raise ValueError(
+                "Unsupported skip_type {}, need to be 'logSNR' or 'time_uniform' or 'time_quadratic'".format(skip_type))
+
+    def get_orders_and_timesteps_for_singlestep_solver(self, steps, order, skip_type, t_T, t_0, device):
+        """
+        Get the order of each step for sampling by the singlestep DPM-Solver.
+        We combine both DPM-Solver-1,2,3 to use all the function evaluations, which is named as "DPM-Solver-fast".
+        Given a fixed number of function evaluations by `steps`, the sampling procedure by DPM-Solver-fast is:
+            - If order == 1:
+                We take `steps` of DPM-Solver-1 (i.e. DDIM).
+            - If order == 2:
+                - Denote K = (steps // 2). We take K or (K + 1) intermediate time steps for sampling.
+                - If steps % 2 == 0, we use K steps of DPM-Solver-2.
+                - If steps % 2 == 1, we use K steps of DPM-Solver-2 and 1 step of DPM-Solver-1.
+            - If order == 3:
+                - Denote K = (steps // 3 + 1). We take K intermediate time steps for sampling.
+                - If steps % 3 == 0, we use (K - 2) steps of DPM-Solver-3, and 1 step of DPM-Solver-2 and 1 step of DPM-Solver-1.
+                - If steps % 3 == 1, we use (K - 1) steps of DPM-Solver-3 and 1 step of DPM-Solver-1.
+                - If steps % 3 == 2, we use (K - 1) steps of DPM-Solver-3 and 1 step of DPM-Solver-2.
+        ============================================
+        Args:
+            order: A `int`. The max order for the solver (2 or 3).
+            steps: A `int`. The total number of function evaluations (NFE).
+            skip_type: A `str`. The type for the spacing of the time steps. We support three types:
+                - 'logSNR': uniform logSNR for the time steps.
+                - 'time_uniform': uniform time for the time steps. (**Recommended for high-resolutional data**.)
+                - 'time_quadratic': quadratic time for the time steps. (Used in DDIM for low-resolutional data.)
+            t_T: A `float`. The starting time of the sampling (default is T).
+            t_0: A `float`. The ending time of the sampling (default is epsilon).
+            device: A torch device.
+        Returns:
+            orders: A list of the solver order of each step.
+        """
+        if order == 3:
+            K = steps // 3 + 1
+            if steps % 3 == 0:
+                orders = [3, ] * (K - 2) + [2, 1]
+            elif steps % 3 == 1:
+                orders = [3, ] * (K - 1) + [1]
+            else:
+                orders = [3, ] * (K - 1) + [2]
+        elif order == 2:
+            if steps % 2 == 0:
+                K = steps // 2
+                orders = [2, ] * K
+            else:
+                K = steps // 2 + 1
+                orders = [2, ] * (K - 1) + [1]
+        elif order == 1:
+            K = 1
+            orders = [1, ] * steps
+        else:
+            raise ValueError("'order' must be '1' or '2' or '3'.")
+        if skip_type == 'logSNR':
+            # To reproduce the results in DPM-Solver paper
+            timesteps_outer = self.get_time_steps(skip_type, t_T, t_0, K, device)
+        else:
+            timesteps_outer = self.get_time_steps(skip_type, t_T, t_0, steps, device)[
+                torch.cumsum(torch.tensor([0, ] + orders)).to(device)]
+        return timesteps_outer, orders
+
+    def denoise_to_zero_fn(self, x, s):
+        """
+        Denoise at the final step, which is equivalent to solve the ODE from lambda_s to infty by first-order discretization.
+        """
+        return self.data_prediction_fn(x, s)
+
+    def dpm_solver_first_update(self, x, s, t, model_s=None, return_intermediate=False):
+        """
+        DPM-Solver-1 (equivalent to DDIM) from time `s` to time `t`.
+        Args:
+            x: A pytorch tensor. The initial value at time `s`.
+            s: A pytorch tensor. The starting time, with the shape (x.shape[0],).
+            t: A pytorch tensor. The ending time, with the shape (x.shape[0],).
+            model_s: A pytorch tensor. The model function evaluated at time `s`.
+                If `model_s` is None, we evaluate the model by `x` and `s`; otherwise we directly use it.
+            return_intermediate: A `bool`. If true, also return the model value at time `s`.
+        Returns:
+            x_t: A pytorch tensor. The approximated solution at time `t`.
+        """
+        ns = self.noise_schedule
+        dims = x.dim()
+        lambda_s, lambda_t = ns.marginal_lambda(s), ns.marginal_lambda(t)
+        h = lambda_t - lambda_s
+        log_alpha_s, log_alpha_t = ns.marginal_log_mean_coeff(s), ns.marginal_log_mean_coeff(t)
+        sigma_s, sigma_t = ns.marginal_std(s), ns.marginal_std(t)
+        alpha_t = torch.exp(log_alpha_t)
+
+        if self.predict_x0:
+            phi_1 = torch.expm1(-h)
+            if model_s is None:
+                model_s = self.model_fn(x, s)
+            x_t = (
+                    expand_dims(sigma_t / sigma_s, dims) * x
+                    - expand_dims(alpha_t * phi_1, dims) * model_s
+            )
+            if return_intermediate:
+                return x_t, {'model_s': model_s}
+            else:
+                return x_t
+        else:
+            phi_1 = torch.expm1(h)
+            if model_s is None:
+                model_s = self.model_fn(x, s)
+            x_t = (
+                    expand_dims(torch.exp(log_alpha_t - log_alpha_s), dims) * x
+                    - expand_dims(sigma_t * phi_1, dims) * model_s
+            )
+            if return_intermediate:
+                return x_t, {'model_s': model_s}
+            else:
+                return x_t
+
+    def singlestep_dpm_solver_second_update(self, x, s, t, r1=0.5, model_s=None, return_intermediate=False,
+                                            solver_type='dpm_solver'):
+        """
+        Singlestep solver DPM-Solver-2 from time `s` to time `t`.
+        Args:
+            x: A pytorch tensor. The initial value at time `s`.
+            s: A pytorch tensor. The starting time, with the shape (x.shape[0],).
+            t: A pytorch tensor. The ending time, with the shape (x.shape[0],).
+            r1: A `float`. The hyperparameter of the second-order solver.
+            model_s: A pytorch tensor. The model function evaluated at time `s`.
+                If `model_s` is None, we evaluate the model by `x` and `s`; otherwise we directly use it.
+            return_intermediate: A `bool`. If true, also return the model value at time `s` and `s1` (the intermediate time).
+            solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers.
+                The type slightly impacts the performance. We recommend to use 'dpm_solver' type.
+        Returns:
+            x_t: A pytorch tensor. The approximated solution at time `t`.
+        """
+        if solver_type not in ['dpm_solver', 'taylor']:
+            raise ValueError("'solver_type' must be either 'dpm_solver' or 'taylor', got {}".format(solver_type))
+        if r1 is None:
+            r1 = 0.5
+        ns = self.noise_schedule
+        dims = x.dim()
+        lambda_s, lambda_t = ns.marginal_lambda(s), ns.marginal_lambda(t)
+        h = lambda_t - lambda_s
+        lambda_s1 = lambda_s + r1 * h
+        s1 = ns.inverse_lambda(lambda_s1)
+        log_alpha_s, log_alpha_s1, log_alpha_t = ns.marginal_log_mean_coeff(s), ns.marginal_log_mean_coeff(
+            s1), ns.marginal_log_mean_coeff(t)
+        sigma_s, sigma_s1, sigma_t = ns.marginal_std(s), ns.marginal_std(s1), ns.marginal_std(t)
+        alpha_s1, alpha_t = torch.exp(log_alpha_s1), torch.exp(log_alpha_t)
+
+        if self.predict_x0:
+            phi_11 = torch.expm1(-r1 * h)
+            phi_1 = torch.expm1(-h)
+
+            if model_s is None:
+                model_s = self.model_fn(x, s)
+            x_s1 = (
+                    expand_dims(sigma_s1 / sigma_s, dims) * x
+                    - expand_dims(alpha_s1 * phi_11, dims) * model_s
+            )
+            model_s1 = self.model_fn(x_s1, s1)
+            if solver_type == 'dpm_solver':
+                x_t = (
+                        expand_dims(sigma_t / sigma_s, dims) * x
+                        - expand_dims(alpha_t * phi_1, dims) * model_s
+                        - (0.5 / r1) * expand_dims(alpha_t * phi_1, dims) * (model_s1 - model_s)
+                )
+            elif solver_type == 'taylor':
+                x_t = (
+                        expand_dims(sigma_t / sigma_s, dims) * x
+                        - expand_dims(alpha_t * phi_1, dims) * model_s
+                        + (1. / r1) * expand_dims(alpha_t * ((torch.exp(-h) - 1.) / h + 1.), dims) * (
+                                    model_s1 - model_s)
+                )
+        else:
+            phi_11 = torch.expm1(r1 * h)
+            phi_1 = torch.expm1(h)
+
+            if model_s is None:
+                model_s = self.model_fn(x, s)
+            x_s1 = (
+                    expand_dims(torch.exp(log_alpha_s1 - log_alpha_s), dims) * x
+                    - expand_dims(sigma_s1 * phi_11, dims) * model_s
+            )
+            model_s1 = self.model_fn(x_s1, s1)
+            if solver_type == 'dpm_solver':
+                x_t = (
+                        expand_dims(torch.exp(log_alpha_t - log_alpha_s), dims) * x
+                        - expand_dims(sigma_t * phi_1, dims) * model_s
+                        - (0.5 / r1) * expand_dims(sigma_t * phi_1, dims) * (model_s1 - model_s)
+                )
+            elif solver_type == 'taylor':
+                x_t = (
+                        expand_dims(torch.exp(log_alpha_t - log_alpha_s), dims) * x
+                        - expand_dims(sigma_t * phi_1, dims) * model_s
+                        - (1. / r1) * expand_dims(sigma_t * ((torch.exp(h) - 1.) / h - 1.), dims) * (model_s1 - model_s)
+                )
+        if return_intermediate:
+            return x_t, {'model_s': model_s, 'model_s1': model_s1}
+        else:
+            return x_t
+
+    def singlestep_dpm_solver_third_update(self, x, s, t, r1=1. / 3., r2=2. / 3., model_s=None, model_s1=None,
+                                           return_intermediate=False, solver_type='dpm_solver'):
+        """
+        Singlestep solver DPM-Solver-3 from time `s` to time `t`.
+        Args:
+            x: A pytorch tensor. The initial value at time `s`.
+            s: A pytorch tensor. The starting time, with the shape (x.shape[0],).
+            t: A pytorch tensor. The ending time, with the shape (x.shape[0],).
+            r1: A `float`. The hyperparameter of the third-order solver.
+            r2: A `float`. The hyperparameter of the third-order solver.
+            model_s: A pytorch tensor. The model function evaluated at time `s`.
+                If `model_s` is None, we evaluate the model by `x` and `s`; otherwise we directly use it.
+            model_s1: A pytorch tensor. The model function evaluated at time `s1` (the intermediate time given by `r1`).
+                If `model_s1` is None, we evaluate the model at `s1`; otherwise we directly use it.
+            return_intermediate: A `bool`. If true, also return the model value at time `s`, `s1` and `s2` (the intermediate times).
+            solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers.
+                The type slightly impacts the performance. We recommend to use 'dpm_solver' type.
+        Returns:
+            x_t: A pytorch tensor. The approximated solution at time `t`.
+        """
+        if solver_type not in ['dpm_solver', 'taylor']:
+            raise ValueError("'solver_type' must be either 'dpm_solver' or 'taylor', got {}".format(solver_type))
+        if r1 is None:
+            r1 = 1. / 3.
+        if r2 is None:
+            r2 = 2. / 3.
+        ns = self.noise_schedule
+        dims = x.dim()
+        lambda_s, lambda_t = ns.marginal_lambda(s), ns.marginal_lambda(t)
+        h = lambda_t - lambda_s
+        lambda_s1 = lambda_s + r1 * h
+        lambda_s2 = lambda_s + r2 * h
+        s1 = ns.inverse_lambda(lambda_s1)
+        s2 = ns.inverse_lambda(lambda_s2)
+        log_alpha_s, log_alpha_s1, log_alpha_s2, log_alpha_t = ns.marginal_log_mean_coeff(
+            s), ns.marginal_log_mean_coeff(s1), ns.marginal_log_mean_coeff(s2), ns.marginal_log_mean_coeff(t)
+        sigma_s, sigma_s1, sigma_s2, sigma_t = ns.marginal_std(s), ns.marginal_std(s1), ns.marginal_std(
+            s2), ns.marginal_std(t)
+        alpha_s1, alpha_s2, alpha_t = torch.exp(log_alpha_s1), torch.exp(log_alpha_s2), torch.exp(log_alpha_t)
+
+        if self.predict_x0:
+            phi_11 = torch.expm1(-r1 * h)
+            phi_12 = torch.expm1(-r2 * h)
+            phi_1 = torch.expm1(-h)
+            phi_22 = torch.expm1(-r2 * h) / (r2 * h) + 1.
+            phi_2 = phi_1 / h + 1.
+            phi_3 = phi_2 / h - 0.5
+
+            if model_s is None:
+                model_s = self.model_fn(x, s)
+            if model_s1 is None:
+                x_s1 = (
+                        expand_dims(sigma_s1 / sigma_s, dims) * x
+                        - expand_dims(alpha_s1 * phi_11, dims) * model_s
+                )
+                model_s1 = self.model_fn(x_s1, s1)
+            x_s2 = (
+                    expand_dims(sigma_s2 / sigma_s, dims) * x
+                    - expand_dims(alpha_s2 * phi_12, dims) * model_s
+                    + r2 / r1 * expand_dims(alpha_s2 * phi_22, dims) * (model_s1 - model_s)
+            )
+            model_s2 = self.model_fn(x_s2, s2)
+            if solver_type == 'dpm_solver':
+                x_t = (
+                        expand_dims(sigma_t / sigma_s, dims) * x
+                        - expand_dims(alpha_t * phi_1, dims) * model_s
+                        + (1. / r2) * expand_dims(alpha_t * phi_2, dims) * (model_s2 - model_s)
+                )
+            elif solver_type == 'taylor':
+                D1_0 = (1. / r1) * (model_s1 - model_s)
+                D1_1 = (1. / r2) * (model_s2 - model_s)
+                D1 = (r2 * D1_0 - r1 * D1_1) / (r2 - r1)
+                D2 = 2. * (D1_1 - D1_0) / (r2 - r1)
+                x_t = (
+                        expand_dims(sigma_t / sigma_s, dims) * x
+                        - expand_dims(alpha_t * phi_1, dims) * model_s
+                        + expand_dims(alpha_t * phi_2, dims) * D1
+                        - expand_dims(alpha_t * phi_3, dims) * D2
+                )
+        else:
+            phi_11 = torch.expm1(r1 * h)
+            phi_12 = torch.expm1(r2 * h)
+            phi_1 = torch.expm1(h)
+            phi_22 = torch.expm1(r2 * h) / (r2 * h) - 1.
+            phi_2 = phi_1 / h - 1.
+            phi_3 = phi_2 / h - 0.5
+
+            if model_s is None:
+                model_s = self.model_fn(x, s)
+            if model_s1 is None:
+                x_s1 = (
+                        expand_dims(torch.exp(log_alpha_s1 - log_alpha_s), dims) * x
+                        - expand_dims(sigma_s1 * phi_11, dims) * model_s
+                )
+                model_s1 = self.model_fn(x_s1, s1)
+            x_s2 = (
+                    expand_dims(torch.exp(log_alpha_s2 - log_alpha_s), dims) * x
+                    - expand_dims(sigma_s2 * phi_12, dims) * model_s
+                    - r2 / r1 * expand_dims(sigma_s2 * phi_22, dims) * (model_s1 - model_s)
+            )
+            model_s2 = self.model_fn(x_s2, s2)
+            if solver_type == 'dpm_solver':
+                x_t = (
+                        expand_dims(torch.exp(log_alpha_t - log_alpha_s), dims) * x
+                        - expand_dims(sigma_t * phi_1, dims) * model_s
+                        - (1. / r2) * expand_dims(sigma_t * phi_2, dims) * (model_s2 - model_s)
+                )
+            elif solver_type == 'taylor':
+                D1_0 = (1. / r1) * (model_s1 - model_s)
+                D1_1 = (1. / r2) * (model_s2 - model_s)
+                D1 = (r2 * D1_0 - r1 * D1_1) / (r2 - r1)
+                D2 = 2. * (D1_1 - D1_0) / (r2 - r1)
+                x_t = (
+                        expand_dims(torch.exp(log_alpha_t - log_alpha_s), dims) * x
+                        - expand_dims(sigma_t * phi_1, dims) * model_s
+                        - expand_dims(sigma_t * phi_2, dims) * D1
+                        - expand_dims(sigma_t * phi_3, dims) * D2
+                )
+
+        if return_intermediate:
+            return x_t, {'model_s': model_s, 'model_s1': model_s1, 'model_s2': model_s2}
+        else:
+            return x_t
+
+    def multistep_dpm_solver_second_update(self, x, model_prev_list, t_prev_list, t, solver_type="dpm_solver"):
+        """
+        Multistep solver DPM-Solver-2 from time `t_prev_list[-1]` to time `t`.
+        Args:
+            x: A pytorch tensor. The initial value at time `s`.
+            model_prev_list: A list of pytorch tensor. The previous computed model values.
+            t_prev_list: A list of pytorch tensor. The previous times, each time has the shape (x.shape[0],)
+            t: A pytorch tensor. The ending time, with the shape (x.shape[0],).
+            solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers.
+                The type slightly impacts the performance. We recommend to use 'dpm_solver' type.
+        Returns:
+            x_t: A pytorch tensor. The approximated solution at time `t`.
+        """
+        if solver_type not in ['dpm_solver', 'taylor']:
+            raise ValueError("'solver_type' must be either 'dpm_solver' or 'taylor', got {}".format(solver_type))
+        ns = self.noise_schedule
+        dims = x.dim()
+        model_prev_1, model_prev_0 = model_prev_list
+        t_prev_1, t_prev_0 = t_prev_list
+        lambda_prev_1, lambda_prev_0, lambda_t = ns.marginal_lambda(t_prev_1), ns.marginal_lambda(
+            t_prev_0), ns.marginal_lambda(t)
+        log_alpha_prev_0, log_alpha_t = ns.marginal_log_mean_coeff(t_prev_0), ns.marginal_log_mean_coeff(t)
+        sigma_prev_0, sigma_t = ns.marginal_std(t_prev_0), ns.marginal_std(t)
+        alpha_t = torch.exp(log_alpha_t)
+
+        h_0 = lambda_prev_0 - lambda_prev_1
+        h = lambda_t - lambda_prev_0
+        r0 = h_0 / h
+        D1_0 = expand_dims(1. / r0, dims) * (model_prev_0 - model_prev_1)
+        if self.predict_x0:
+            if solver_type == 'dpm_solver':
+                x_t = (
+                        expand_dims(sigma_t / sigma_prev_0, dims) * x
+                        - expand_dims(alpha_t * (torch.exp(-h) - 1.), dims) * model_prev_0
+                        - 0.5 * expand_dims(alpha_t * (torch.exp(-h) - 1.), dims) * D1_0
+                )
+            elif solver_type == 'taylor':
+                x_t = (
+                        expand_dims(sigma_t / sigma_prev_0, dims) * x
+                        - expand_dims(alpha_t * (torch.exp(-h) - 1.), dims) * model_prev_0
+                        + expand_dims(alpha_t * ((torch.exp(-h) - 1.) / h + 1.), dims) * D1_0
+                )
+        else:
+            if solver_type == 'dpm_solver':
+                x_t = (
+                        expand_dims(torch.exp(log_alpha_t - log_alpha_prev_0), dims) * x
+                        - expand_dims(sigma_t * (torch.exp(h) - 1.), dims) * model_prev_0
+                        - 0.5 * expand_dims(sigma_t * (torch.exp(h) - 1.), dims) * D1_0
+                )
+            elif solver_type == 'taylor':
+                x_t = (
+                        expand_dims(torch.exp(log_alpha_t - log_alpha_prev_0), dims) * x
+                        - expand_dims(sigma_t * (torch.exp(h) - 1.), dims) * model_prev_0
+                        - expand_dims(sigma_t * ((torch.exp(h) - 1.) / h - 1.), dims) * D1_0
+                )
+        return x_t
+
+    def multistep_dpm_solver_third_update(self, x, model_prev_list, t_prev_list, t, solver_type='dpm_solver'):
+        """
+        Multistep solver DPM-Solver-3 from time `t_prev_list[-1]` to time `t`.
+        Args:
+            x: A pytorch tensor. The initial value at time `s`.
+            model_prev_list: A list of pytorch tensor. The previous computed model values.
+            t_prev_list: A list of pytorch tensor. The previous times, each time has the shape (x.shape[0],)
+            t: A pytorch tensor. The ending time, with the shape (x.shape[0],).
+            solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers.
+                The type slightly impacts the performance. We recommend to use 'dpm_solver' type.
+        Returns:
+            x_t: A pytorch tensor. The approximated solution at time `t`.
+        """
+        ns = self.noise_schedule
+        dims = x.dim()
+        model_prev_2, model_prev_1, model_prev_0 = model_prev_list
+        t_prev_2, t_prev_1, t_prev_0 = t_prev_list
+        lambda_prev_2, lambda_prev_1, lambda_prev_0, lambda_t = ns.marginal_lambda(t_prev_2), ns.marginal_lambda(
+            t_prev_1), ns.marginal_lambda(t_prev_0), ns.marginal_lambda(t)
+        log_alpha_prev_0, log_alpha_t = ns.marginal_log_mean_coeff(t_prev_0), ns.marginal_log_mean_coeff(t)
+        sigma_prev_0, sigma_t = ns.marginal_std(t_prev_0), ns.marginal_std(t)
+        alpha_t = torch.exp(log_alpha_t)
+
+        h_1 = lambda_prev_1 - lambda_prev_2
+        h_0 = lambda_prev_0 - lambda_prev_1
+        h = lambda_t - lambda_prev_0
+        r0, r1 = h_0 / h, h_1 / h
+        D1_0 = expand_dims(1. / r0, dims) * (model_prev_0 - model_prev_1)
+        D1_1 = expand_dims(1. / r1, dims) * (model_prev_1 - model_prev_2)
+        D1 = D1_0 + expand_dims(r0 / (r0 + r1), dims) * (D1_0 - D1_1)
+        D2 = expand_dims(1. / (r0 + r1), dims) * (D1_0 - D1_1)
+        if self.predict_x0:
+            x_t = (
+                    expand_dims(sigma_t / sigma_prev_0, dims) * x
+                    - expand_dims(alpha_t * (torch.exp(-h) - 1.), dims) * model_prev_0
+                    + expand_dims(alpha_t * ((torch.exp(-h) - 1.) / h + 1.), dims) * D1
+                    - expand_dims(alpha_t * ((torch.exp(-h) - 1. + h) / h ** 2 - 0.5), dims) * D2
+            )
+        else:
+            x_t = (
+                    expand_dims(torch.exp(log_alpha_t - log_alpha_prev_0), dims) * x
+                    - expand_dims(sigma_t * (torch.exp(h) - 1.), dims) * model_prev_0
+                    - expand_dims(sigma_t * ((torch.exp(h) - 1.) / h - 1.), dims) * D1
+                    - expand_dims(sigma_t * ((torch.exp(h) - 1. - h) / h ** 2 - 0.5), dims) * D2
+            )
+        return x_t
+
+    def singlestep_dpm_solver_update(self, x, s, t, order, return_intermediate=False, solver_type='dpm_solver', r1=None,
+                                     r2=None):
+        """
+        Singlestep DPM-Solver with the order `order` from time `s` to time `t`.
+        Args:
+            x: A pytorch tensor. The initial value at time `s`.
+            s: A pytorch tensor. The starting time, with the shape (x.shape[0],).
+            t: A pytorch tensor. The ending time, with the shape (x.shape[0],).
+            order: A `int`. The order of DPM-Solver. We only support order == 1 or 2 or 3.
+            return_intermediate: A `bool`. If true, also return the model value at time `s`, `s1` and `s2` (the intermediate times).
+            solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers.
+                The type slightly impacts the performance. We recommend to use 'dpm_solver' type.
+            r1: A `float`. The hyperparameter of the second-order or third-order solver.
+            r2: A `float`. The hyperparameter of the third-order solver.
+        Returns:
+            x_t: A pytorch tensor. The approximated solution at time `t`.
+        """
+        if order == 1:
+            return self.dpm_solver_first_update(x, s, t, return_intermediate=return_intermediate)
+        elif order == 2:
+            return self.singlestep_dpm_solver_second_update(x, s, t, return_intermediate=return_intermediate,
+                                                            solver_type=solver_type, r1=r1)
+        elif order == 3:
+            return self.singlestep_dpm_solver_third_update(x, s, t, return_intermediate=return_intermediate,
+                                                           solver_type=solver_type, r1=r1, r2=r2)
+        else:
+            raise ValueError("Solver order must be 1 or 2 or 3, got {}".format(order))
+
+    def multistep_dpm_solver_update(self, x, model_prev_list, t_prev_list, t, order, solver_type='dpm_solver'):
+        """
+        Multistep DPM-Solver with the order `order` from time `t_prev_list[-1]` to time `t`.
+        Args:
+            x: A pytorch tensor. The initial value at time `s`.
+            model_prev_list: A list of pytorch tensor. The previous computed model values.
+            t_prev_list: A list of pytorch tensor. The previous times, each time has the shape (x.shape[0],)
+            t: A pytorch tensor. The ending time, with the shape (x.shape[0],).
+            order: A `int`. The order of DPM-Solver. We only support order == 1 or 2 or 3.
+            solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers.
+                The type slightly impacts the performance. We recommend to use 'dpm_solver' type.
+        Returns:
+            x_t: A pytorch tensor. The approximated solution at time `t`.
+        """
+        if order == 1:
+            return self.dpm_solver_first_update(x, t_prev_list[-1], t, model_s=model_prev_list[-1])
+        elif order == 2:
+            return self.multistep_dpm_solver_second_update(x, model_prev_list, t_prev_list, t, solver_type=solver_type)
+        elif order == 3:
+            return self.multistep_dpm_solver_third_update(x, model_prev_list, t_prev_list, t, solver_type=solver_type)
+        else:
+            raise ValueError("Solver order must be 1 or 2 or 3, got {}".format(order))
+
+    def dpm_solver_adaptive(self, x, order, t_T, t_0, h_init=0.05, atol=0.0078, rtol=0.05, theta=0.9, t_err=1e-5,
+                            solver_type='dpm_solver'):
+        """
+        The adaptive step size solver based on singlestep DPM-Solver.
+        Args:
+            x: A pytorch tensor. The initial value at time `t_T`.
+            order: A `int`. The (higher) order of the solver. We only support order == 2 or 3.
+            t_T: A `float`. The starting time of the sampling (default is T).
+            t_0: A `float`. The ending time of the sampling (default is epsilon).
+            h_init: A `float`. The initial step size (for logSNR).
+            atol: A `float`. The absolute tolerance of the solver. For image data, the default setting is 0.0078, followed [1].
+            rtol: A `float`. The relative tolerance of the solver. The default setting is 0.05.
+            theta: A `float`. The safety hyperparameter for adapting the step size. The default setting is 0.9, followed [1].
+            t_err: A `float`. The tolerance for the time. We solve the diffusion ODE until the absolute error between the
+                current time and `t_0` is less than `t_err`. The default setting is 1e-5.
+            solver_type: either 'dpm_solver' or 'taylor'. The type for the high-order solvers.
+                The type slightly impacts the performance. We recommend to use 'dpm_solver' type.
+        Returns:
+            x_0: A pytorch tensor. The approximated solution at time `t_0`.
+        [1] A. Jolicoeur-Martineau, K. Li, R. Piché-Taillefer, T. Kachman, and I. Mitliagkas, "Gotta go fast when generating data with score-based models," arXiv preprint arXiv:2105.14080, 2021.
+        """
+        ns = self.noise_schedule
+        s = t_T * torch.ones((x.shape[0],)).to(x)
+        lambda_s = ns.marginal_lambda(s)
+        lambda_0 = ns.marginal_lambda(t_0 * torch.ones_like(s).to(x))
+        h = h_init * torch.ones_like(s).to(x)
+        x_prev = x
+        nfe = 0
+        if order == 2:
+            r1 = 0.5
+            lower_update = lambda x, s, t: self.dpm_solver_first_update(x, s, t, return_intermediate=True)
+            higher_update = lambda x, s, t, **kwargs: self.singlestep_dpm_solver_second_update(x, s, t, r1=r1,
+                                                                                               solver_type=solver_type,
+                                                                                               **kwargs)
+        elif order == 3:
+            r1, r2 = 1. / 3., 2. / 3.
+            lower_update = lambda x, s, t: self.singlestep_dpm_solver_second_update(x, s, t, r1=r1,
+                                                                                    return_intermediate=True,
+                                                                                    solver_type=solver_type)
+            higher_update = lambda x, s, t, **kwargs: self.singlestep_dpm_solver_third_update(x, s, t, r1=r1, r2=r2,
+                                                                                              solver_type=solver_type,
+                                                                                              **kwargs)
+        else:
+            raise ValueError("For adaptive step size solver, order must be 2 or 3, got {}".format(order))
+        while torch.abs((s - t_0)).mean() > t_err:
+            t = ns.inverse_lambda(lambda_s + h)
+            x_lower, lower_noise_kwargs = lower_update(x, s, t)
+            x_higher = higher_update(x, s, t, **lower_noise_kwargs)
+            delta = torch.max(torch.ones_like(x).to(x) * atol, rtol * torch.max(torch.abs(x_lower), torch.abs(x_prev)))
+            norm_fn = lambda v: torch.sqrt(torch.square(v.reshape((v.shape[0], -1))).mean(dim=-1, keepdim=True))
+            E = norm_fn((x_higher - x_lower) / delta).max()
+            if torch.all(E <= 1.):
+                x = x_higher
+                s = t
+                x_prev = x_lower
+                lambda_s = ns.marginal_lambda(s)
+            h = torch.min(theta * h * torch.float_power(E, -1. / order).float(), lambda_0 - lambda_s)
+            nfe += order
+        print('adaptive solver nfe', nfe)
+        return x
+
+    def sample(self, x, steps=20, t_start=None, t_end=None, order=3, skip_type='time_uniform',
+               method='singlestep', lower_order_final=True, denoise_to_zero=False, solver_type='dpm_solver',
+               atol=0.0078, rtol=0.05,
+               ):
+        """
+        Compute the sample at time `t_end` by DPM-Solver, given the initial `x` at time `t_start`.
+        =====================================================
+        We support the following algorithms for both noise prediction model and data prediction model:
+            - 'singlestep':
+                Singlestep DPM-Solver (i.e. "DPM-Solver-fast" in the paper), which combines different orders of singlestep DPM-Solver.
+                We combine all the singlestep solvers with order <= `order` to use up all the function evaluations (steps).
+                The total number of function evaluations (NFE) == `steps`.
+                Given a fixed NFE == `steps`, the sampling procedure is:
+                    - If `order` == 1:
+                        - Denote K = steps. We use K steps of DPM-Solver-1 (i.e. DDIM).
+                    - If `order` == 2:
+                        - Denote K = (steps // 2) + (steps % 2). We take K intermediate time steps for sampling.
+                        - If steps % 2 == 0, we use K steps of singlestep DPM-Solver-2.
+                        - If steps % 2 == 1, we use (K - 1) steps of singlestep DPM-Solver-2 and 1 step of DPM-Solver-1.
+                    - If `order` == 3:
+                        - Denote K = (steps // 3 + 1). We take K intermediate time steps for sampling.
+                        - If steps % 3 == 0, we use (K - 2) steps of singlestep DPM-Solver-3, and 1 step of singlestep DPM-Solver-2 and 1 step of DPM-Solver-1.
+                        - If steps % 3 == 1, we use (K - 1) steps of singlestep DPM-Solver-3 and 1 step of DPM-Solver-1.
+                        - If steps % 3 == 2, we use (K - 1) steps of singlestep DPM-Solver-3 and 1 step of singlestep DPM-Solver-2.
+            - 'multistep':
+                Multistep DPM-Solver with the order of `order`. The total number of function evaluations (NFE) == `steps`.
+                We initialize the first `order` values by lower order multistep solvers.
+                Given a fixed NFE == `steps`, the sampling procedure is:
+                    Denote K = steps.
+                    - If `order` == 1:
+                        - We use K steps of DPM-Solver-1 (i.e. DDIM).
+                    - If `order` == 2:
+                        - We firstly use 1 step of DPM-Solver-1, then use (K - 1) step of multistep DPM-Solver-2.
+                    - If `order` == 3:
+                        - We firstly use 1 step of DPM-Solver-1, then 1 step of multistep DPM-Solver-2, then (K - 2) step of multistep DPM-Solver-3.
+            - 'singlestep_fixed':
+                Fixed order singlestep DPM-Solver (i.e. DPM-Solver-1 or singlestep DPM-Solver-2 or singlestep DPM-Solver-3).
+                We use singlestep DPM-Solver-`order` for `order`=1 or 2 or 3, with total [`steps` // `order`] * `order` NFE.
+            - 'adaptive':
+                Adaptive step size DPM-Solver (i.e. "DPM-Solver-12" and "DPM-Solver-23" in the paper).
+                We ignore `steps` and use adaptive step size DPM-Solver with a higher order of `order`.
+                You can adjust the absolute tolerance `atol` and the relative tolerance `rtol` to balance the computatation costs
+                (NFE) and the sample quality.
+                    - If `order` == 2, we use DPM-Solver-12 which combines DPM-Solver-1 and singlestep DPM-Solver-2.
+                    - If `order` == 3, we use DPM-Solver-23 which combines singlestep DPM-Solver-2 and singlestep DPM-Solver-3.
+        =====================================================
+        Some advices for choosing the algorithm:
+            - For **unconditional sampling** or **guided sampling with small guidance scale** by DPMs:
+                Use singlestep DPM-Solver ("DPM-Solver-fast" in the paper) with `order = 3`.
+                e.g.
+                    >>> dpm_solver = DPM_Solver(model_fn, noise_schedule, predict_x0=False)
+                    >>> x_sample = dpm_solver.sample(x, steps=steps, t_start=t_start, t_end=t_end, order=3,
+                            skip_type='time_uniform', method='singlestep')
+            - For **guided sampling with large guidance scale** by DPMs:
+                Use multistep DPM-Solver with `predict_x0 = True` and `order = 2`.
+                e.g.
+                    >>> dpm_solver = DPM_Solver(model_fn, noise_schedule, predict_x0=True)
+                    >>> x_sample = dpm_solver.sample(x, steps=steps, t_start=t_start, t_end=t_end, order=2,
+                            skip_type='time_uniform', method='multistep')
+        We support three types of `skip_type`:
+            - 'logSNR': uniform logSNR for the time steps. **Recommended for low-resolutional images**
+            - 'time_uniform': uniform time for the time steps. **Recommended for high-resolutional images**.
+            - 'time_quadratic': quadratic time for the time steps.
+        =====================================================
+        Args:
+            x: A pytorch tensor. The initial value at time `t_start`
+                e.g. if `t_start` == T, then `x` is a sample from the standard normal distribution.
+            steps: A `int`. The total number of function evaluations (NFE).
+            t_start: A `float`. The starting time of the sampling.
+                If `T` is None, we use self.noise_schedule.T (default is 1.0).
+            t_end: A `float`. The ending time of the sampling.
+                If `t_end` is None, we use 1. / self.noise_schedule.total_N.
+                e.g. if total_N == 1000, we have `t_end` == 1e-3.
+                For discrete-time DPMs:
+                    - We recommend `t_end` == 1. / self.noise_schedule.total_N.
+                For continuous-time DPMs:
+                    - We recommend `t_end` == 1e-3 when `steps` <= 15; and `t_end` == 1e-4 when `steps` > 15.
+            order: A `int`. The order of DPM-Solver.
+            skip_type: A `str`. The type for the spacing of the time steps. 'time_uniform' or 'logSNR' or 'time_quadratic'.
+            method: A `str`. The method for sampling. 'singlestep' or 'multistep' or 'singlestep_fixed' or 'adaptive'.
+            denoise_to_zero: A `bool`. Whether to denoise to time 0 at the final step.
+                Default is `False`. If `denoise_to_zero` is `True`, the total NFE is (`steps` + 1).
+                This trick is firstly proposed by DDPM (https://arxiv.org/abs/2006.11239) and
+                score_sde (https://arxiv.org/abs/2011.13456). Such trick can improve the FID
+                for diffusion models sampling by diffusion SDEs for low-resolutional images
+                (such as CIFAR-10). However, we observed that such trick does not matter for
+                high-resolutional images. As it needs an additional NFE, we do not recommend
+                it for high-resolutional images.
+            lower_order_final: A `bool`. Whether to use lower order solvers at the final steps.
+                Only valid for `method=multistep` and `steps < 15`. We empirically find that
+                this trick is a key to stabilizing the sampling by DPM-Solver with very few steps
+                (especially for steps <= 10). So we recommend to set it to be `True`.
+            solver_type: A `str`. The taylor expansion type for the solver. `dpm_solver` or `taylor`. We recommend `dpm_solver`.
+            atol: A `float`. The absolute tolerance of the adaptive step size solver. Valid when `method` == 'adaptive'.
+            rtol: A `float`. The relative tolerance of the adaptive step size solver. Valid when `method` == 'adaptive'.
+        Returns:
+            x_end: A pytorch tensor. The approximated solution at time `t_end`.
+        """
+        t_0 = 1. / self.noise_schedule.total_N if t_end is None else t_end
+        t_T = self.noise_schedule.T if t_start is None else t_start
+        device = x.device
+        if method == 'adaptive':
+            with torch.no_grad():
+                x = self.dpm_solver_adaptive(x, order=order, t_T=t_T, t_0=t_0, atol=atol, rtol=rtol,
+                                             solver_type=solver_type)
+        elif method == 'multistep':
+            assert steps >= order
+            timesteps = self.get_time_steps(skip_type=skip_type, t_T=t_T, t_0=t_0, N=steps, device=device)
+            assert timesteps.shape[0] - 1 == steps
+            with torch.no_grad():
+                vec_t = timesteps[0].expand((x.shape[0]))
+                model_prev_list = [self.model_fn(x, vec_t)]
+                t_prev_list = [vec_t]
+                # Init the first `order` values by lower order multistep DPM-Solver.
+                for init_order in tqdm(range(1, order), desc="DPM init order"):
+                    vec_t = timesteps[init_order].expand(x.shape[0])
+                    x = self.multistep_dpm_solver_update(x, model_prev_list, t_prev_list, vec_t, init_order,
+                                                         solver_type=solver_type)
+                    model_prev_list.append(self.model_fn(x, vec_t))
+                    t_prev_list.append(vec_t)
+                # Compute the remaining values by `order`-th order multistep DPM-Solver.
+                for step in tqdm(range(order, steps + 1), desc="DPM multistep"):
+                    vec_t = timesteps[step].expand(x.shape[0])
+                    if lower_order_final and steps < 15:
+                        step_order = min(order, steps + 1 - step)
+                    else:
+                        step_order = order
+                    x = self.multistep_dpm_solver_update(x, model_prev_list, t_prev_list, vec_t, step_order,
+                                                         solver_type=solver_type)
+                    for i in range(order - 1):
+                        t_prev_list[i] = t_prev_list[i + 1]
+                        model_prev_list[i] = model_prev_list[i + 1]
+                    t_prev_list[-1] = vec_t
+                    # We do not need to evaluate the final model value.
+                    if step < steps:
+                        model_prev_list[-1] = self.model_fn(x, vec_t)
+        elif method in ['singlestep', 'singlestep_fixed']:
+            if method == 'singlestep':
+                timesteps_outer, orders = self.get_orders_and_timesteps_for_singlestep_solver(steps=steps, order=order,
+                                                                                              skip_type=skip_type,
+                                                                                              t_T=t_T, t_0=t_0,
+                                                                                              device=device)
+            elif method == 'singlestep_fixed':
+                K = steps // order
+                orders = [order, ] * K
+                timesteps_outer = self.get_time_steps(skip_type=skip_type, t_T=t_T, t_0=t_0, N=K, device=device)
+            for i, order in enumerate(orders):
+                t_T_inner, t_0_inner = timesteps_outer[i], timesteps_outer[i + 1]
+                timesteps_inner = self.get_time_steps(skip_type=skip_type, t_T=t_T_inner.item(), t_0=t_0_inner.item(),
+                                                      N=order, device=device)
+                lambda_inner = self.noise_schedule.marginal_lambda(timesteps_inner)
+                vec_s, vec_t = t_T_inner.tile(x.shape[0]), t_0_inner.tile(x.shape[0])
+                h = lambda_inner[-1] - lambda_inner[0]
+                r1 = None if order <= 1 else (lambda_inner[1] - lambda_inner[0]) / h
+                r2 = None if order <= 2 else (lambda_inner[2] - lambda_inner[0]) / h
+                x = self.singlestep_dpm_solver_update(x, vec_s, vec_t, order, solver_type=solver_type, r1=r1, r2=r2)
+        if denoise_to_zero:
+            x = self.denoise_to_zero_fn(x, torch.ones((x.shape[0],)).to(device) * t_0)
+        return x
+
+
+#############################################################
+# other utility functions
+#############################################################
+
+def interpolate_fn(x, xp, yp):
+    """
+    A piecewise linear function y = f(x), using xp and yp as keypoints.
+    We implement f(x) in a differentiable way (i.e. applicable for autograd).
+    The function f(x) is well-defined for all x-axis. (For x beyond the bounds of xp, we use the outmost points of xp to define the linear function.)
+    Args:
+        x: PyTorch tensor with shape [N, C], where N is the batch size, C is the number of channels (we use C = 1 for DPM-Solver).
+        xp: PyTorch tensor with shape [C, K], where K is the number of keypoints.
+        yp: PyTorch tensor with shape [C, K].
+    Returns:
+        The function values f(x), with shape [N, C].
+    """
+    N, K = x.shape[0], xp.shape[1]
+    all_x = torch.cat([x.unsqueeze(2), xp.unsqueeze(0).repeat((N, 1, 1))], dim=2)
+    sorted_all_x, x_indices = torch.sort(all_x, dim=2)
+    x_idx = torch.argmin(x_indices, dim=2)
+    cand_start_idx = x_idx - 1
+    start_idx = torch.where(
+        torch.eq(x_idx, 0),
+        torch.tensor(1, device=x.device),
+        torch.where(
+            torch.eq(x_idx, K), torch.tensor(K - 2, device=x.device), cand_start_idx,
+        ),
+    )
+    end_idx = torch.where(torch.eq(start_idx, cand_start_idx), start_idx + 2, start_idx + 1)
+    start_x = torch.gather(sorted_all_x, dim=2, index=start_idx.unsqueeze(2)).squeeze(2)
+    end_x = torch.gather(sorted_all_x, dim=2, index=end_idx.unsqueeze(2)).squeeze(2)
+    start_idx2 = torch.where(
+        torch.eq(x_idx, 0),
+        torch.tensor(0, device=x.device),
+        torch.where(
+            torch.eq(x_idx, K), torch.tensor(K - 2, device=x.device), cand_start_idx,
+        ),
+    )
+    y_positions_expanded = yp.unsqueeze(0).expand(N, -1, -1)
+    start_y = torch.gather(y_positions_expanded, dim=2, index=start_idx2.unsqueeze(2)).squeeze(2)
+    end_y = torch.gather(y_positions_expanded, dim=2, index=(start_idx2 + 1).unsqueeze(2)).squeeze(2)
+    cand = start_y + (x - start_x) * (end_y - start_y) / (end_x - start_x)
+    return cand
+
+
+def expand_dims(v, dims):
+    """
+    Expand the tensor `v` to the dim `dims`.
+    Args:
+        `v`: a PyTorch tensor with shape [N].
+        `dim`: a `int`.
+    Returns:
+        a PyTorch tensor with shape [N, 1, 1, ..., 1] and the total dimension is `dims`.
+    """
+    return v[(...,) + (None,) * (dims - 1)]

ldm/models/diffusion/dpm_solver/sampler.py

@@ -0,0 +1,87 @@
+"""SAMPLING ONLY."""
+import torch
+
+from .dpm_solver import NoiseScheduleVP, model_wrapper, DPM_Solver
+
+
+MODEL_TYPES = {
+    "eps": "noise",
+    "v": "v"
+}
+
+
+class DPMSolverSampler(object):
+    def __init__(self, model, **kwargs):
+        super().__init__()
+        self.model = model
+        to_torch = lambda x: x.clone().detach().to(torch.float32).to(model.device)
+        self.register_buffer('alphas_cumprod', to_torch(model.alphas_cumprod))
+
+    def register_buffer(self, name, attr):
+        if type(attr) == torch.Tensor:
+            if attr.device != torch.device("cuda"):
+                attr = attr.to(torch.device("cuda"))
+        setattr(self, name, attr)
+
+    @torch.no_grad()
+    def sample(self,
+               S,
+               batch_size,
+               shape,
+               conditioning=None,
+               callback=None,
+               normals_sequence=None,
+               img_callback=None,
+               quantize_x0=False,
+               eta=0.,
+               mask=None,
+               x0=None,
+               temperature=1.,
+               noise_dropout=0.,
+               score_corrector=None,
+               corrector_kwargs=None,
+               verbose=True,
+               x_T=None,
+               log_every_t=100,
+               unconditional_guidance_scale=1.,
+               unconditional_conditioning=None,
+               # this has to come in the same format as the conditioning, # e.g. as encoded tokens, ...
+               **kwargs
+               ):
+        if conditioning is not None:
+            if isinstance(conditioning, dict):
+                cbs = conditioning[list(conditioning.keys())[0]].shape[0]
+                if cbs != batch_size:
+                    print(f"Warning: Got {cbs} conditionings but batch-size is {batch_size}")
+            else:
+                if conditioning.shape[0] != batch_size:
+                    print(f"Warning: Got {conditioning.shape[0]} conditionings but batch-size is {batch_size}")
+
+        # sampling
+        C, H, W = shape
+        size = (batch_size, C, H, W)
+
+        print(f'Data shape for DPM-Solver sampling is {size}, sampling steps {S}')
+
+        device = self.model.betas.device
+        if x_T is None:
+            img = torch.randn(size, device=device)
+        else:
+            img = x_T
+
+        ns = NoiseScheduleVP('discrete', alphas_cumprod=self.alphas_cumprod)
+
+        model_fn = model_wrapper(
+            lambda x, t, c: self.model.apply_model(x, t, c),
+            ns,
+            model_type=MODEL_TYPES[self.model.parameterization],
+            guidance_type="classifier-free",
+            condition=conditioning,
+            unconditional_condition=unconditional_conditioning,
+            guidance_scale=unconditional_guidance_scale,
+        )
+
+        dpm_solver = DPM_Solver(model_fn, ns, predict_x0=True, thresholding=False)
+        x = dpm_solver.sample(img, steps=S, skip_type="time_uniform", method="multistep", order=2, lower_order_final=True)
+
+        return x.to(device), None

ldm/models/diffusion/plms.py

@@ -0,0 +1,244 @@
+"""SAMPLING ONLY."""
+
+import torch
+import numpy as np
+from tqdm import tqdm
+from functools import partial
+
+from ldm.modules.diffusionmodules.util import make_ddim_sampling_parameters, make_ddim_timesteps, noise_like
+from ldm.models.diffusion.sampling_util import norm_thresholding
+
+
+class PLMSSampler(object):
+    def __init__(self, model, schedule="linear", **kwargs):
+        super().__init__()
+        self.model = model
+        self.ddpm_num_timesteps = model.num_timesteps
+        self.schedule = schedule
+
+    def register_buffer(self, name, attr):
+        if type(attr) == torch.Tensor:
+            if attr.device != torch.device("cuda"):
+                attr = attr.to(torch.device("cuda"))
+        setattr(self, name, attr)
+
+    def make_schedule(self, ddim_num_steps, ddim_discretize="uniform", ddim_eta=0., verbose=True):
+        if ddim_eta != 0:
+            raise ValueError('ddim_eta must be 0 for PLMS')
+        self.ddim_timesteps = make_ddim_timesteps(ddim_discr_method=ddim_discretize, num_ddim_timesteps=ddim_num_steps,
+                                                  num_ddpm_timesteps=self.ddpm_num_timesteps,verbose=verbose)
+        alphas_cumprod = self.model.alphas_cumprod
+        assert alphas_cumprod.shape[0] == self.ddpm_num_timesteps, 'alphas have to be defined for each timestep'
+        to_torch = lambda x: x.clone().detach().to(torch.float32).to(self.model.device)
+
+        self.register_buffer('betas', to_torch(self.model.betas))
+        self.register_buffer('alphas_cumprod', to_torch(alphas_cumprod))
+        self.register_buffer('alphas_cumprod_prev', to_torch(self.model.alphas_cumprod_prev))
+
+        # calculations for diffusion q(x_t | x_{t-1}) and others
+        self.register_buffer('sqrt_alphas_cumprod', to_torch(np.sqrt(alphas_cumprod.cpu())))
+        self.register_buffer('sqrt_one_minus_alphas_cumprod', to_torch(np.sqrt(1. - alphas_cumprod.cpu())))
+        self.register_buffer('log_one_minus_alphas_cumprod', to_torch(np.log(1. - alphas_cumprod.cpu())))
+        self.register_buffer('sqrt_recip_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod.cpu())))
+        self.register_buffer('sqrt_recipm1_alphas_cumprod', to_torch(np.sqrt(1. / alphas_cumprod.cpu() - 1)))
+
+        # ddim sampling parameters
+        ddim_sigmas, ddim_alphas, ddim_alphas_prev = make_ddim_sampling_parameters(alphacums=alphas_cumprod.cpu(),
+                                                                                   ddim_timesteps=self.ddim_timesteps,
+                                                                                   eta=ddim_eta,verbose=verbose)
+        self.register_buffer('ddim_sigmas', ddim_sigmas)
+        self.register_buffer('ddim_alphas', ddim_alphas)
+        self.register_buffer('ddim_alphas_prev', ddim_alphas_prev)
+        self.register_buffer('ddim_sqrt_one_minus_alphas', np.sqrt(1. - ddim_alphas))
+        sigmas_for_original_sampling_steps = ddim_eta * torch.sqrt(
+            (1 - self.alphas_cumprod_prev) / (1 - self.alphas_cumprod) * (
+                        1 - self.alphas_cumprod / self.alphas_cumprod_prev))
+        self.register_buffer('ddim_sigmas_for_original_num_steps', sigmas_for_original_sampling_steps)
+
+    @torch.no_grad()
+    def sample(self,
+               S,
+               batch_size,
+               shape,
+               conditioning=None,
+               callback=None,
+               normals_sequence=None,
+               img_callback=None,
+               quantize_x0=False,
+               eta=0.,
+               mask=None,
+               x0=None,
+               temperature=1.,
+               noise_dropout=0.,
+               score_corrector=None,
+               corrector_kwargs=None,
+               verbose=True,
+               x_T=None,
+               log_every_t=100,
+               unconditional_guidance_scale=1.,
+               unconditional_conditioning=None,
+               # this has to come in the same format as the conditioning, # e.g. as encoded tokens, ...
+               dynamic_threshold=None,
+               **kwargs
+               ):
+        if conditioning is not None:
+            if isinstance(conditioning, dict):
+                cbs = conditioning[list(conditioning.keys())[0]].shape[0]
+                if cbs != batch_size:
+                    print(f"Warning: Got {cbs} conditionings but batch-size is {batch_size}")
+            else:
+                if conditioning.shape[0] != batch_size:
+                    print(f"Warning: Got {conditioning.shape[0]} conditionings but batch-size is {batch_size}")
+
+        self.make_schedule(ddim_num_steps=S, ddim_eta=eta, verbose=verbose)
+        # sampling
+        C, H, W = shape
+        size = (batch_size, C, H, W)
+        print(f'Data shape for PLMS sampling is {size}')
+
+        samples, intermediates = self.plms_sampling(conditioning, size,
+                                                    callback=callback,
+                                                    img_callback=img_callback,
+                                                    quantize_denoised=quantize_x0,
+                                                    mask=mask, x0=x0,
+                                                    ddim_use_original_steps=False,
+                                                    noise_dropout=noise_dropout,
+                                                    temperature=temperature,
+                                                    score_corrector=score_corrector,
+                                                    corrector_kwargs=corrector_kwargs,
+                                                    x_T=x_T,
+                                                    log_every_t=log_every_t,
+                                                    unconditional_guidance_scale=unconditional_guidance_scale,
+                                                    unconditional_conditioning=unconditional_conditioning,
+                                                    dynamic_threshold=dynamic_threshold,
+                                                    )
+        return samples, intermediates
+
+    @torch.no_grad()
+    def plms_sampling(self, cond, shape,
+                      x_T=None, ddim_use_original_steps=False,
+                      callback=None, timesteps=None, quantize_denoised=False,
+                      mask=None, x0=None, img_callback=None, log_every_t=100,
+                      temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None,
+                      unconditional_guidance_scale=1., unconditional_conditioning=None,
+                      dynamic_threshold=None):
+        device = self.model.betas.device
+        b = shape[0]
+        if x_T is None:
+            img = torch.randn(shape, device=device)
+        else:
+            img = x_T
+
+        if timesteps is None:
+            timesteps = self.ddpm_num_timesteps if ddim_use_original_steps else self.ddim_timesteps
+        elif timesteps is not None and not ddim_use_original_steps:
+            subset_end = int(min(timesteps / self.ddim_timesteps.shape[0], 1) * self.ddim_timesteps.shape[0]) - 1
+            timesteps = self.ddim_timesteps[:subset_end]
+
+        intermediates = {'x_inter': [img], 'pred_x0': [img]}
+        time_range = list(reversed(range(0,timesteps))) if ddim_use_original_steps else np.flip(timesteps)
+        total_steps = timesteps if ddim_use_original_steps else timesteps.shape[0]
+        print(f"Running PLMS Sampling with {total_steps} timesteps")
+
+        iterator = tqdm(time_range, desc='PLMS Sampler', total=total_steps)
+        old_eps = []
+
+        for i, step in enumerate(iterator):
+            index = total_steps - i - 1
+            ts = torch.full((b,), step, device=device, dtype=torch.long)
+            ts_next = torch.full((b,), time_range[min(i + 1, len(time_range) - 1)], device=device, dtype=torch.long)
+
+            if mask is not None:
+                assert x0 is not None
+                img_orig = self.model.q_sample(x0, ts)  # TODO: deterministic forward pass?
+                img = img_orig * mask + (1. - mask) * img
+
+            outs = self.p_sample_plms(img, cond, ts, index=index, use_original_steps=ddim_use_original_steps,
+                                      quantize_denoised=quantize_denoised, temperature=temperature,
+                                      noise_dropout=noise_dropout, score_corrector=score_corrector,
+                                      corrector_kwargs=corrector_kwargs,
+                                      unconditional_guidance_scale=unconditional_guidance_scale,
+                                      unconditional_conditioning=unconditional_conditioning,
+                                      old_eps=old_eps, t_next=ts_next,
+                                      dynamic_threshold=dynamic_threshold)
+            img, pred_x0, e_t = outs
+            old_eps.append(e_t)
+            if len(old_eps) >= 4:
+                old_eps.pop(0)
+            if callback: callback(i)
+            if img_callback: img_callback(pred_x0, i)
+
+            if index % log_every_t == 0 or index == total_steps - 1:
+                intermediates['x_inter'].append(img)
+                intermediates['pred_x0'].append(pred_x0)
+
+        return img, intermediates
+
+    @torch.no_grad()
+    def p_sample_plms(self, x, c, t, index, repeat_noise=False, use_original_steps=False, quantize_denoised=False,
+                      temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None,
+                      unconditional_guidance_scale=1., unconditional_conditioning=None, old_eps=None, t_next=None,
+                      dynamic_threshold=None):
+        b, *_, device = *x.shape, x.device
+
+        def get_model_output(x, t):
+            if unconditional_conditioning is None or unconditional_guidance_scale == 1.:
+                e_t = self.model.apply_model(x, t, c)
+            else:
+                x_in = torch.cat([x] * 2)
+                t_in = torch.cat([t] * 2)
+                c_in = torch.cat([unconditional_conditioning, c])
+                e_t_uncond, e_t = self.model.apply_model(x_in, t_in, c_in).chunk(2)
+                e_t = e_t_uncond + unconditional_guidance_scale * (e_t - e_t_uncond)
+
+            if score_corrector is not None:
+                assert self.model.parameterization == "eps"
+                e_t = score_corrector.modify_score(self.model, e_t, x, t, c, **corrector_kwargs)
+
+            return e_t
+
+        alphas = self.model.alphas_cumprod if use_original_steps else self.ddim_alphas
+        alphas_prev = self.model.alphas_cumprod_prev if use_original_steps else self.ddim_alphas_prev
+        sqrt_one_minus_alphas = self.model.sqrt_one_minus_alphas_cumprod if use_original_steps else self.ddim_sqrt_one_minus_alphas
+        sigmas = self.model.ddim_sigmas_for_original_num_steps if use_original_steps else self.ddim_sigmas
+
+        def get_x_prev_and_pred_x0(e_t, index):
+            # select parameters corresponding to the currently considered timestep
+            a_t = torch.full((b, 1, 1, 1), alphas[index], device=device)
+            a_prev = torch.full((b, 1, 1, 1), alphas_prev[index], device=device)
+            sigma_t = torch.full((b, 1, 1, 1), sigmas[index], device=device)
+            sqrt_one_minus_at = torch.full((b, 1, 1, 1), sqrt_one_minus_alphas[index],device=device)
+
+            # current prediction for x_0
+            pred_x0 = (x - sqrt_one_minus_at * e_t) / a_t.sqrt()
+            if quantize_denoised:
+                pred_x0, _, *_ = self.model.first_stage_model.quantize(pred_x0)
+            if dynamic_threshold is not None:
+                pred_x0 = norm_thresholding(pred_x0, dynamic_threshold)
+            # direction pointing to x_t
+            dir_xt = (1. - a_prev - sigma_t**2).sqrt() * e_t
+            noise = sigma_t * noise_like(x.shape, device, repeat_noise) * temperature
+            if noise_dropout > 0.:
+                noise = torch.nn.functional.dropout(noise, p=noise_dropout)
+            x_prev = a_prev.sqrt() * pred_x0 + dir_xt + noise
+            return x_prev, pred_x0
+
+        e_t = get_model_output(x, t)
+        if len(old_eps) == 0:
+            # Pseudo Improved Euler (2nd order)
+            x_prev, pred_x0 = get_x_prev_and_pred_x0(e_t, index)
+            e_t_next = get_model_output(x_prev, t_next)
+            e_t_prime = (e_t + e_t_next) / 2
+        elif len(old_eps) == 1:
+            # 2nd order Pseudo Linear Multistep (Adams-Bashforth)
+            e_t_prime = (3 * e_t - old_eps[-1]) / 2
+        elif len(old_eps) == 2:
+            # 3nd order Pseudo Linear Multistep (Adams-Bashforth)
+            e_t_prime = (23 * e_t - 16 * old_eps[-1] + 5 * old_eps[-2]) / 12
+        elif len(old_eps) >= 3:
+            # 4nd order Pseudo Linear Multistep (Adams-Bashforth)
+            e_t_prime = (55 * e_t - 59 * old_eps[-1] + 37 * old_eps[-2] - 9 * old_eps[-3]) / 24
+
+        x_prev, pred_x0 = get_x_prev_and_pred_x0(e_t_prime, index)
+
+        return x_prev, pred_x0, e_t

ldm/models/diffusion/sampling_util.py

@@ -0,0 +1,22 @@
+import torch
+import numpy as np
+
+
+def append_dims(x, target_dims):
+    """Appends dimensions to the end of a tensor until it has target_dims dimensions.
+    From https://github.com/crowsonkb/k-diffusion/blob/master/k_diffusion/utils.py"""
+    dims_to_append = target_dims - x.ndim
+    if dims_to_append < 0:
+        raise ValueError(f'input has {x.ndim} dims but target_dims is {target_dims}, which is less')
+    return x[(...,) + (None,) * dims_to_append]
+
+
+def norm_thresholding(x0, value):
+    s = append_dims(x0.pow(2).flatten(1).mean(1).sqrt().clamp(min=value), x0.ndim)
+    return x0 * (value / s)
+
+
+def spatial_norm_thresholding(x0, value):
+    # b c h w
+    s = x0.pow(2).mean(1, keepdim=True).sqrt().clamp(min=value)
+    return x0 * (value / s)

ldm/modules/attention.py

@@ -0,0 +1,341 @@
+from inspect import isfunction
+import math
+import torch
+import torch.nn.functional as F
+from torch import nn, einsum
+from einops import rearrange, repeat
+from typing import Optional, Any
+
+from ldm.modules.diffusionmodules.util import checkpoint
+
+
+try:
+    import xformers
+    import xformers.ops
+    XFORMERS_IS_AVAILBLE = True
+except:
+    XFORMERS_IS_AVAILBLE = False
+
+# CrossAttn precision handling
+import os
+_ATTN_PRECISION = os.environ.get("ATTN_PRECISION", "fp32")
+
+def exists(val):
+    return val is not None
+
+
+def uniq(arr):
+    return{el: True for el in arr}.keys()
+
+
+def default(val, d):
+    if exists(val):
+        return val
+    return d() if isfunction(d) else d
+
+
+def max_neg_value(t):
+    return -torch.finfo(t.dtype).max
+
+
+def init_(tensor):
+    dim = tensor.shape[-1]
+    std = 1 / math.sqrt(dim)
+    tensor.uniform_(-std, std)
+    return tensor
+
+
+# feedforward
+class GEGLU(nn.Module):
+    def __init__(self, dim_in, dim_out):
+        super().__init__()
+        self.proj = nn.Linear(dim_in, dim_out * 2)
+
+    def forward(self, x):
+        x, gate = self.proj(x).chunk(2, dim=-1)
+        return x * F.gelu(gate)
+
+
+class FeedForward(nn.Module):
+    def __init__(self, dim, dim_out=None, mult=4, glu=False, dropout=0.):
+        super().__init__()
+        inner_dim = int(dim * mult)
+        dim_out = default(dim_out, dim)
+        project_in = nn.Sequential(
+            nn.Linear(dim, inner_dim),
+            nn.GELU()
+        ) if not glu else GEGLU(dim, inner_dim)
+
+        self.net = nn.Sequential(
+            project_in,
+            nn.Dropout(dropout),
+            nn.Linear(inner_dim, dim_out)
+        )
+
+    def forward(self, x):
+        return self.net(x)
+
+
+def zero_module(module):
+    """
+    Zero out the parameters of a module and return it.
+    """
+    for p in module.parameters():
+        p.detach().zero_()
+    return module
+
+
+def Normalize(in_channels):
+    return torch.nn.GroupNorm(num_groups=32, num_channels=in_channels, eps=1e-6, affine=True)
+
+
+class SpatialSelfAttention(nn.Module):
+    def __init__(self, in_channels):
+        super().__init__()
+        self.in_channels = in_channels
+
+        self.norm = Normalize(in_channels)
+        self.q = torch.nn.Conv2d(in_channels,
+                                 in_channels,
+                                 kernel_size=1,
+                                 stride=1,
+                                 padding=0)
+        self.k = torch.nn.Conv2d(in_channels,
+                                 in_channels,
+                                 kernel_size=1,
+                                 stride=1,
+                                 padding=0)
+        self.v = torch.nn.Conv2d(in_channels,
+                                 in_channels,
+                                 kernel_size=1,
+                                 stride=1,
+                                 padding=0)
+        self.proj_out = torch.nn.Conv2d(in_channels,
+                                        in_channels,
+                                        kernel_size=1,
+                                        stride=1,
+                                        padding=0)
+
+    def forward(self, x):
+        h_ = x
+        h_ = self.norm(h_)
+        q = self.q(h_)
+        k = self.k(h_)
+        v = self.v(h_)
+
+        # compute attention
+        b,c,h,w = q.shape
+        q = rearrange(q, 'b c h w -> b (h w) c')
+        k = rearrange(k, 'b c h w -> b c (h w)')
+        w_ = torch.einsum('bij,bjk->bik', q, k)
+
+        w_ = w_ * (int(c)**(-0.5))
+        w_ = torch.nn.functional.softmax(w_, dim=2)
+
+        # attend to values
+        v = rearrange(v, 'b c h w -> b c (h w)')
+        w_ = rearrange(w_, 'b i j -> b j i')
+        h_ = torch.einsum('bij,bjk->bik', v, w_)
+        h_ = rearrange(h_, 'b c (h w) -> b c h w', h=h)
+        h_ = self.proj_out(h_)
+
+        return x+h_
+
+
+class CrossAttention(nn.Module):
+    def __init__(self, query_dim, context_dim=None, heads=8, dim_head=64, dropout=0.):
+        super().__init__()
+        inner_dim = dim_head * heads
+        context_dim = default(context_dim, query_dim)
+
+        self.scale = dim_head ** -0.5
+        self.heads = heads
+
+        self.to_q = nn.Linear(query_dim, inner_dim, bias=False)
+        self.to_k = nn.Linear(context_dim, inner_dim, bias=False)
+        self.to_v = nn.Linear(context_dim, inner_dim, bias=False)
+
+        self.to_out = nn.Sequential(
+            nn.Linear(inner_dim, query_dim),
+            nn.Dropout(dropout)
+        )
+
+    def forward(self, x, context=None, mask=None):
+        h = self.heads
+
+        q = self.to_q(x)
+        context = default(context, x)
+        k = self.to_k(context)
+        v = self.to_v(context)
+
+        q, k, v = map(lambda t: rearrange(t, 'b n (h d) -> (b h) n d', h=h), (q, k, v))
+
+        # force cast to fp32 to avoid overflowing
+        if _ATTN_PRECISION =="fp32":
+            with torch.autocast(enabled=False, device_type = 'cuda'):
+                q, k = q.float(), k.float()
+                sim = einsum('b i d, b j d -> b i j', q, k) * self.scale
+        else:
+            sim = einsum('b i d, b j d -> b i j', q, k) * self.scale
+        
+        del q, k
+    
+        if exists(mask):
+            mask = rearrange(mask, 'b ... -> b (...)')
+            max_neg_value = -torch.finfo(sim.dtype).max
+            mask = repeat(mask, 'b j -> (b h) () j', h=h)
+            sim.masked_fill_(~mask, max_neg_value)
+
+        # attention, what we cannot get enough of
+        sim = sim.softmax(dim=-1)
+
+        out = einsum('b i j, b j d -> b i d', sim, v)
+        out = rearrange(out, '(b h) n d -> b n (h d)', h=h)
+        return self.to_out(out)
+
+
+class MemoryEfficientCrossAttention(nn.Module):
+    # https://github.com/MatthieuTPHR/diffusers/blob/d80b531ff8060ec1ea982b65a1b8df70f73aa67c/src/diffusers/models/attention.py#L223
+    def __init__(self, query_dim, context_dim=None, heads=8, dim_head=64, dropout=0.0):
+        super().__init__()
+        print(f"Setting up {self.__class__.__name__}. Query dim is {query_dim}, context_dim is {context_dim} and using "
+              f"{heads} heads.")
+        inner_dim = dim_head * heads
+        context_dim = default(context_dim, query_dim)
+
+        self.heads = heads
+        self.dim_head = dim_head
+
+        self.to_q = nn.Linear(query_dim, inner_dim, bias=False)
+        self.to_k = nn.Linear(context_dim, inner_dim, bias=False)
+        self.to_v = nn.Linear(context_dim, inner_dim, bias=False)
+
+        self.to_out = nn.Sequential(nn.Linear(inner_dim, query_dim), nn.Dropout(dropout))
+        self.attention_op: Optional[Any] = None
+
+    def forward(self, x, context=None, mask=None):
+        q = self.to_q(x)
+        context = default(context, x)
+        k = self.to_k(context)
+        v = self.to_v(context)
+
+        b, _, _ = q.shape
+        q, k, v = map(
+            lambda t: t.unsqueeze(3)
+            .reshape(b, t.shape[1], self.heads, self.dim_head)
+            .permute(0, 2, 1, 3)
+            .reshape(b * self.heads, t.shape[1], self.dim_head)
+            .contiguous(),
+            (q, k, v),
+        )
+
+        # actually compute the attention, what we cannot get enough of
+        out = xformers.ops.memory_efficient_attention(q, k, v, attn_bias=None, op=self.attention_op)
+
+        if exists(mask):
+            raise NotImplementedError
+        out = (
+            out.unsqueeze(0)
+            .reshape(b, self.heads, out.shape[1], self.dim_head)
+            .permute(0, 2, 1, 3)
+            .reshape(b, out.shape[1], self.heads * self.dim_head)
+        )
+        return self.to_out(out)
+
+
+class BasicTransformerBlock(nn.Module):
+    ATTENTION_MODES = {
+        "softmax": CrossAttention,  # vanilla attention
+        "softmax-xformers": MemoryEfficientCrossAttention
+    }
+    def __init__(self, dim, n_heads, d_head, dropout=0., context_dim=None, gated_ff=True, checkpoint=True,
+                 disable_self_attn=False):
+        super().__init__()
+        attn_mode = "softmax-xformers" if XFORMERS_IS_AVAILBLE else "softmax"
+        assert attn_mode in self.ATTENTION_MODES
+        attn_cls = self.ATTENTION_MODES[attn_mode]
+        self.disable_self_attn = disable_self_attn
+        self.attn1 = attn_cls(query_dim=dim, heads=n_heads, dim_head=d_head, dropout=dropout,
+                              context_dim=context_dim if self.disable_self_attn else None)  # is a self-attention if not self.disable_self_attn
+        self.ff = FeedForward(dim, dropout=dropout, glu=gated_ff)
+        self.attn2 = attn_cls(query_dim=dim, context_dim=context_dim,
+                              heads=n_heads, dim_head=d_head, dropout=dropout)  # is self-attn if context is none
+        self.norm1 = nn.LayerNorm(dim)
+        self.norm2 = nn.LayerNorm(dim)
+        self.norm3 = nn.LayerNorm(dim)
+        self.checkpoint = checkpoint
+
+    def forward(self, x, context=None):
+        return checkpoint(self._forward, (x, context), self.parameters(), self.checkpoint)
+
+    def _forward(self, x, context=None):
+        x = self.attn1(self.norm1(x), context=context if self.disable_self_attn else None) + x
+        x = self.attn2(self.norm2(x), context=context) + x
+        x = self.ff(self.norm3(x)) + x
+        return x
+
+
+class SpatialTransformer(nn.Module):
+    """
+    Transformer block for image-like data.
+    First, project the input (aka embedding)
+    and reshape to b, t, d.
+    Then apply standard transformer action.
+    Finally, reshape to image
+    NEW: use_linear for more efficiency instead of the 1x1 convs
+    """
+    def __init__(self, in_channels, n_heads, d_head,
+                 depth=1, dropout=0., context_dim=None,
+                 disable_self_attn=False, use_linear=False,
+                 use_checkpoint=True):
+        super().__init__()
+        if exists(context_dim) and not isinstance(context_dim, list):
+            context_dim = [context_dim]
+        self.in_channels = in_channels
+        inner_dim = n_heads * d_head
+        self.norm = Normalize(in_channels)
+        if not use_linear:
+            self.proj_in = nn.Conv2d(in_channels,
+                                     inner_dim,
+                                     kernel_size=1,
+                                     stride=1,
+                                     padding=0)
+        else:
+            self.proj_in = nn.Linear(in_channels, inner_dim)
+
+        self.transformer_blocks = nn.ModuleList(
+            [BasicTransformerBlock(inner_dim, n_heads, d_head, dropout=dropout, context_dim=context_dim[d],
+                                   disable_self_attn=disable_self_attn, checkpoint=use_checkpoint)
+                for d in range(depth)]
+        )
+        if not use_linear:
+            self.proj_out = zero_module(nn.Conv2d(inner_dim,
+                                                  in_channels,
+                                                  kernel_size=1,
+                                                  stride=1,
+                                                  padding=0))
+        else:
+            self.proj_out = zero_module(nn.Linear(in_channels, inner_dim))
+        self.use_linear = use_linear
+
+    def forward(self, x, context=None):
+        # note: if no context is given, cross-attention defaults to self-attention
+        if not isinstance(context, list):
+            context = [context]
+        b, c, h, w = x.shape
+        x_in = x
+        x = self.norm(x)
+        if not self.use_linear:
+            x = self.proj_in(x)
+        x = rearrange(x, 'b c h w -> b (h w) c').contiguous()
+        if self.use_linear:
+            x = self.proj_in(x)
+        for i, block in enumerate(self.transformer_blocks):
+            x = block(x, context=context[i])
+        if self.use_linear:
+            x = self.proj_out(x)
+        x = rearrange(x, 'b (h w) c -> b c h w', h=h, w=w).contiguous()
+        if not self.use_linear:
+            x = self.proj_out(x)
+        return x + x_in
+