Update app.py

app.py

@@ -1,7 +1,89 @@
+import os
+import sys
 
+sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+import numpy as np
 import streamlit as st
+from PIL import Image
+
+import clip
+from dalle.models import Dalle
+from dalle.utils.utils import clip_score
+
+
+device = "cpu"
+model = Dalle.from_pretrained("minDALL-E/1.3B")  # This will automatically download the pretrained model.
+model.to(device=device)
+
+model_clip, preprocess_clip = clip.load("ViT-B/32", device=device)
+model_clip.to(device=device)
+
+
+def sample(prompt):
+    # Sampling
+    images = (
+        model.sampling(prompt=prompt, top_k=256, top_p=None, softmax_temperature=1.0, num_candidates=3, device=device)
+        .cpu()
+        .numpy()
+    )
+    images = np.transpose(images, (0, 2, 3, 1))
+
+    # CLIP Re-ranking
+    rank = clip_score(
+        prompt=prompt, images=images, model_clip=model_clip, preprocess_clip=preprocess_clip, device=device
+    )
 
-st.header("DALL·E mini")
+    # Save images
+    images = images[rank]
+    # print(rank, images.shape)
+    pil_images = []
+    for i in range(len(images)):
+        im = Image.fromarray((images[i] * 255).astype(np.uint8))
+        pil_images.append(im)
+
+    # im = Image.fromarray((images[0] * 255).astype(np.uint8))
+    return pil_images
+
+
+st.header("minDALL-E")
 st.subheader("Generate images from text")
 
 prompt = st.text_input("What do you want to see?")
+
+DEBUG = False
+if prompt != "":
+    container = st.empty()
+    container.markdown(
+        f"""
+        <style> p {{ margin:0 }} div {{ margin:0 }} </style>
+        <div data-stale="false" class="element-container css-1e5imcs e1tzin5v1">
+        <div class="stAlert">
+        <div role="alert" data-baseweb="notification" class="st-ae st-af st-ag st-ah st-ai st-aj st-ak st-g3 st-am st-b8 st-ao st-ap st-aq st-ar st-as st-at st-au st-av st-aw st-ax st-ay st-az st-b9 st-b1 st-b2 st-b3 st-b4 st-b5 st-b6">
+        <div class="st-b7">
+        <div class="css-whx05o e13vu3m50">
+        <div data-testid="stMarkdownContainer" class="css-1ekf893 e16nr0p30">
+                <img src="https://raw.githubusercontent.com/borisdayma/dalle-mini/main/app/streamlit/img/loading.gif" width="30"/>
+                Generating predictions for: <b>{prompt}</b>
+        </div>
+        </div>
+        </div>
+        </div>
+        </div>
+        </div>
+        <small><i>Predictions may take up to 40s under high load. Please stand by.</i></small>
+    """,
+        unsafe_allow_html=True,
+    )
+
+    print(f"Getting selections: {prompt}")
+    selected = sample(prompt)
+
+    margin = 0.1  # for better position of zoom in arrow
+    n_columns = 3
+    cols = st.columns([1] + [margin, 1] * (n_columns - 1))
+    for i, img in enumerate(selected):
+        cols[(i % n_columns) * 2].image(img)
+    container.markdown(f"**{prompt}**")
+
+    st.button("Again!", key="again_button")