initial commit

app.py

@@ -0,0 +1,41 @@
+import gradio as gr
+from transformers import CLIPProcessor, CLIPModel
+
+model = CLIPModel.from_pretrained("openai/clip-vit-base-patch16")
+processor = CLIPProcessor.from_pretrained("openai/clip-vit-base-patch16")
+
+
+def calculate_CLIP_score(image, text):
+    words = text.split(";")
+    words = [w.strip() for w in words]
+    words = list(filter(None, words))
+
+    if len(words) == 0:
+        return dict()
+    
+    inputs = processor(text=words, images=image, return_tensors="pt", padding=True)
+    outputs = model(**inputs)
+    logits_per_image = outputs.logits_per_image.detach().numpy()
+
+    results_dict = {
+        label: score / 100.0 
+        for label, score in zip(words, logits_per_image[0])
+    }
+    
+    return results_dict
+
+
+examples = [
+    ["images/two_dogs.jpg", "two dogs playing in the beach; a dog and a dog playing in the beach; beach"],
+    ["images/horse_field.jpg", "horse standing in a field; a field; a horse standing"],
+    ["images/human.jpg", "a man beside a river; a riverbed; a man"]
+]
+
+demo = gr.Interface(
+    fn=calculate_CLIP_score,
+    inputs=["image", "text"],
+    outputs="label",
+    examples=examples,
+)
+
+demo.launch()

requirements.txt

@@ -0,0 +1,4 @@
+torch
+transformers
+gradio
+torchvision