Create app.py

app.py

@@ -0,0 +1,36 @@
+import gradio as gr
+from transformers import AutoTokenizer, AutoModel
+import torch
+
+# Load the tokenizer and model from Hugging Face
+tokenizer = AutoTokenizer.from_pretrained("sentence-transformers/all-MiniLM-L6-v2")
+model = AutoModel.from_pretrained("sentence-transformers/all-MiniLM-L6-v2")
+
+def compute_similarity(text1, text2):
+    # Tokenize the input texts
+    inputs = tokenizer([text1, text2], padding=True, truncation=True, return_tensors='pt')
+    
+    # Get the embeddings
+    with torch.no_grad():
+        outputs = model(**inputs)
+    
+    # Compute the mean pooling for both embeddings
+    embeddings = outputs.last_hidden_state.mean(dim=1)
+    
+    # Compute the cosine similarity
+    similarity = torch.nn.functional.cosine_similarity(embeddings[0], embeddings[1], dim=0)
+    
+    return similarity.item()
+
+# Define the Gradio interface
+iface = gr.Interface(
+    fn=compute_similarity,
+    inputs=[gr.inputs.Textbox(lines=2, placeholder="Enter first sentence here..."), gr.inputs.Textbox(lines=2, placeholder="Enter second sentence here...")],
+    outputs="text",
+    title="Text Similarity Model",
+    description="Compute the similarity between two sentences using a pre-trained Hugging Face model."
+)
+
+# Launch the Gradio app
+if __name__ == "__main__":
+    iface.launch()