Create app.py

app.py

@@ -0,0 +1,40 @@
+import gradio as gr
+import open_clip
+import torch
+import requests
+import numpy as np
+from PIL import Image
+
+shapes = ["leggings", "jogger", 
+          "palazzo", "cargo",
+          "dresspants", "chinos"]
+
+model, preprocess_train, preprocess_val = open_clip.create_model_and_transforms('hf-hub:Marqo/marqo-fashionCLIP')
+tokenizer = open_clip.get_tokenizer('hf-hub:Marqo/marqo-fashionCLIP')
+
+
+shapes_desc = list(map(lambda x: "a " + x + " pants shape", shapes))
+text = tokenizer(shapes_desc)
+
+with torch.no_grad(), torch.cuda.amp.autocast():
+  text_features = model.encode_text(text)
+  text_features /= text_features.norm(dim=-1, keepdim=True)
+
+def predict(inp):
+  image = preprocess_val(inp).unsqueeze(0)
+
+  with torch.no_grad(), torch.cuda.amp.autocast():
+      image_features = model.encode_image(image)
+      image_features /= image_features.norm(dim=-1, keepdim=True)
+
+      text_probs = (100 * image_features @ text_features.T).softmax(dim=-1)
+      confidences = {shapes[i]: float(text_probs[0, i]) for i in range(6)}
+  return confidences
+
+
+gr.Interface(fn=predict,
+             inputs=gr.Image(type="pil"),
+             outputs=gr.Label(num_top_classes=6),
+             examples=["imgs/cargo.jpg", "imgs/palazzo.jpg",
+                      "imgs/leggings.jpg", "imgs/jogger.jpg",
+                      "imgs/chinos.jpg", "imgs/dresspants.jpg"]).launch(share=True)