Update app.py

app.py

@@ -1,50 +1,60 @@
-import gradio as gr
-import torch
-from torchvision import models, transforms
-from PIL import Image
-import requests
-from io import BytesIO
+from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
+from qwen_vl_utils import process_vision_info
 
-# Load pre-trained MobileNetV2 model (you can choose another model as needed)
-model = models.mobilenet_v2(pretrained=True)
-model.eval()
+# default: Load the model on the available device(s)
+model = Qwen2VLForConditionalGeneration.from_pretrained(
+    "Qwen/Qwen2-VL-2B-Instruct", torch_dtype="auto", device_map="auto"
+)
 
-# Define the image transformation (resize, normalization)
-transform = transforms.Compose([
-    transforms.Resize(256),
-    transforms.CenterCrop(224),
-    transforms.ToTensor(),
-    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
-])
+# We recommend enabling flash_attention_2 for better acceleration and memory saving, especially in multi-image and video scenarios.
+# model = Qwen2VLForConditionalGeneration.from_pretrained(
+#     "Qwen/Qwen2-VL-2B-Instruct",
+#     torch_dtype=torch.bfloat16,
+#     attn_implementation="flash_attention_2",
+#     device_map="auto",
+# )
 
-# Download the ImageNet class labels (you can replace this with your own if needed)
-LABELS_URL = "https://storage.googleapis.com/download.tensorflow.org/data/imagenet_class_index.json"
-class_idx = requests.get(LABELS_URL).json()
-idx2label = [class_idx[str(k)][1] for k in range(len(class_idx))]
+# default processer
+processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-2B-Instruct")
 
-# Function to perform image inference
-def predict_image(image):
-    image = Image.open(BytesIO(image)).convert("RGB")
-    image = transform(image).unsqueeze(0)
-    
-    # Perform inference
-    with torch.no_grad():
-        output = model(image)
-    
-    # Get the predicted label
-    _, predicted_class = torch.max(output, 1)
-    label = idx2label[predicted_class.item()]
-    return label
+# The default range for the number of visual tokens per image in the model is 4-16384. You can set min_pixels and max_pixels according to your needs, such as a token count range of 256-1280, to balance speed and memory usage.
+# min_pixels = 256*28*28
+# max_pixels = 1280*28*28
+# processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-2B-Instruct", min_pixels=min_pixels, max_pixels=max_pixels)
 
-# Gradio interface
-with gr.Blocks() as demo:
-    gr.Markdown("### Image Classification with MobileNetV2")
-    
-    with gr.Row():
-        image_input = gr.Image(type="bytes", label="Upload Image")
-        result_output = gr.Textbox(label="Predicted Label")
-        
-    # Endpoint for inference
-    gr.Interface(fn=predict_image, inputs=image_input, outputs=result_output, api_name="/predict_image")
+messages = [
+    {
+        "role": "user",
+        "content": [
+            {
+                "type": "image",
+                "image": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg",
+            },
+            {"type": "text", "text": "Describe this image."},
+        ],
+    }
+]
 
-demo.launch(debug=True)
+# Preparation for inference
+text = processor.apply_chat_template(
+    messages, tokenize=False, add_generation_prompt=True
+)
+image_inputs, video_inputs = process_vision_info(messages)
+inputs = processor(
+    text=[text],
+    images=image_inputs,
+    videos=video_inputs,
+    padding=True,
+    return_tensors="pt",
+)
+inputs = inputs.to("cuda")
+
+# Inference: Generation of the output
+generated_ids = model.generate(**inputs, max_new_tokens=128)
+generated_ids_trimmed = [
+    out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
+]
+output_text = processor.batch_decode(
+    generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
+)
+print(output_text)