Update app.py

app.py

@@ -1,18 +1,33 @@
-!pip install git+https://github.com/openai/CLIP.git
-
-from PIL import Image
-import requests
-
 from transformers import CLIPProcessor, CLIPModel
 
 model = CLIPModel.from_pretrained("openai/clip-vit-base-patch32")
 processor = CLIPProcessor.from_pretrained("openai/clip-vit-base-patch32")
 
-url = "http://images.cocodataset.org/val2017/000000039769.jpg"
-image = Image.open(requests.get(url, stream=True).raw)
 
-inputs = processor(text=["a photo of a cat", "a photo of a dog"], images=image, return_tensors="pt", padding=True)
+def inference(input_img, captions):
+    captions_list = captions.split(",")
+    #url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+    #image = Image.open(requests.get(url, stream=True).raw)
+    inputs = processor(text=captions_list, images=input_img, return_tensors="pt", padding=True)
+    outputs = model(**inputs)
+    logits_per_image = outputs.logits_per_image  # this is the image-text similarity score
+    probs = logits_per_image.softmax(dim=1) 
+    probabilities_percentages = ', '.join(['{:.2f}%'.format(prob.item() * 100) for prob in probs[0]])
+    return probabilities_percentages
+
+title = "TSAI S18 Assignment: Use a pretrained CLIP model and give a demo on its workig"
+description = "A simple Gradio interface that accepts an image and some captions, and gives a score as to how much the caption describes the image "
+ 
+examples = [["cats.jpg","a photo of a cat, a photo of a dog"]
+           ]
+
+demo = gr.Interface(
+    inference, 
+    inputs = [gr.Image(shape=(416, 416), label="Input Image"), gr.Textbox(placeholder="Enter different captions for image, separated by comma")], 
+    outputs = [gr.Textbox(label="Probability score of captions")],
+    title = title,
+    description = description,
+    examples = examples,
+)
 
-outputs = model(**inputs)
-logits_per_image = outputs.logits_per_image  # this is the image-text similarity score
-probs = logits_per_image.softmax(dim=1)  # we can take the softmax to get the label probabilities
+demo.launch()