update temp

app.py

@@ -1,7 +1,77 @@
 import gradio as gr
+from datasets import load_dataset
+from transformers import CLIPTokenizerFast, CLIPProcessor, CLIPModel
+import torch
+from tqdm.auto import tqdm
+import numpy as np
+
+device = 'cpu' # 'cuda' if torch.cuda.is_available() else "cpu"
+model_id = 'openai/clip-vit-base-patch32'
+model = CLIPModel.from_pretrained(model_id).to(device)
+tokenizer = CLIPTokenizerFast.from_pretrained(model_id)
+processor = CLIPProcessor.from_pretrained(model_id)
+
+
 
 def greet(name):
     return "Hello " + name + "!!"
 
-iface = gr.Interface(fn=greet, inputs="text", outputs="text")
-iface.launch(inline = False )
+def load_data():
+    global imagenette
+    imagenette = load_dataset(
+        'frgfm/imagenette',
+        'full_size',
+        split = 'train',
+        ignore_verifications = False # set to True if seeing splits Error
+    )
+    return imagenette
+
+def embedding_input(text_input):
+    token_input = tokenizer(text_input, return_tensors = "pt")
+    text_emb = model.get_text_features(**token_input.to(device))
+    return text_emb
+
+def embedding_img():
+    global images
+    load_data()
+    sample_idx= np.random.randint(0, len(imagenette)+1, 100).tolist()
+    images = [imagenette[i]['image'] for i in sample_idx]
+    batch_sie = 5
+    image_arr = None
+    for i in tqdm(range(0, len(images), batch_sie)):
+        batch = images[i:i+batch_sie]
+
+        batch = processor(
+            text = None,
+            images = batch,
+            return_tensors= 'pt',
+            padding = True
+        )['pixel_values'].to(device)
+        batch_emb = model.get_image_features(pixel_values = batch)
+        batch_emb = batch_emb.squeeze(0)
+        batch_emb = batch_emb.cpu().detach().numpy()
+        if image_arr is None:
+            image_arr = batch_emb
+            
+        else:
+            image_arr = np.concatenate((image_arr, batch_emb), axis = 0)
+        return image_arr
+
+def norm_val(text_input):
+    image_arr = embedding_img()
+    text_emb = embedding_input(text_input)
+
+    image_arr = (image_arr.T / np.linalg.norm(image_arr, axis = 1)).T
+    text_emb = text_emb.cpu().detach().numpy()
+    scores = np.dot(text_emb, image_arr.T)
+    top_k = 1
+    idx = np.argsort(-scores[0])[:top_k]
+    return images[idx[0]]
+        
+
+
+
+
+if __name__ == "__main__":
+    iface = gr.Interface(fn=norm_val, inputs="text", outputs="image")
+    iface.launch(inline = False )

requirements.txt

@@ -0,0 +1,7 @@
+torch
+gradio
+numpy
+pandas
+datasets
+tqdm
+transformers