Create app.py

app.py

@@ -0,0 +1,49 @@
+import gradio as gr
+import torch
+import pandas as pd
+from sentence_transformers import SentenceTransformer
+
+# embedding model
+model = SentenceTransformer("all-MiniLM-L6-v2") 
+
+#supplier data
+df = pd.read_csv("suppliers.csv")
+
+supplier_names = df["Supplier Name"].tolist()
+supplier_texts = df["Capabilities"].fillna("").tolist()
+
+# Precompute supplier embeddings
+supplier_embeddings = model.encode(supplier_texts, convert_to_tensor=True, normalize_embeddings=True)
+
+def get_top_supplier(event_description, top_k=5):
+    # Get embedding for the input event
+    event_embedding = model.encode(event_description, convert_to_tensor=True, normalize_embeddings=True).unsqueeze(0)
+    
+    # Compute cosine similarity
+    scores = torch.nn.functional.cosine_similarity(event_embedding, supplier_embeddings)
+    
+    # Find top K matches
+    top_indices = torch.topk(scores, k=top_k).indices.tolist()
+    
+    # Format results
+    results = []
+    for idx in top_indices:
+        results.append({
+            "Supplier Name": supplier_names[idx],
+            "Match Score": round(scores[idx].item(), 4),
+            "Capabilities": supplier_texts[idx]
+        })
+    
+    return pd.DataFrame(results)
+
+# Gradio UI
+demo = gr.Interface(
+    fn=get_top_supplier,
+    inputs=gr.Textbox(lines=4, placeholder="Describe your use case here...", label="Use Case Description"),
+    outputs=gr.Dataframe(label="Top Matching Suppliers"),
+    title="Supplier Matching App",
+    description="Enter a use case or event description to find the most relevant suppliers based on their capabilities."
+)
+
+if __name__ == "__main__":
+    demo.launch()