Update app.py

app.py

@@ -3,47 +3,66 @@ import torch
 import pandas as pd
 from sentence_transformers import SentenceTransformer
 
-# embedding model
-model = SentenceTransformer("all-MiniLM-L6-v2") 
+# Model
+model = SentenceTransformer("all-MiniLM-L6-v2")
 
-#supplier data
-df = pd.read_excel("SupplierList.xlsx")
+# Global state for reusability
+supplier_df = None
+supplier_embeddings = None
 
-supplier_names = df["Supplier Name"].tolist()
-supplier_texts = df["Capability"].fillna("").tolist()
+def load_and_process_supplier_data(file):
+    global supplier_df, supplier_embeddings
 
-# Precompute supplier embeddings
-supplier_embeddings = model.encode(supplier_texts, convert_to_tensor=True, normalize_embeddings=True)
+    supplier_df = pd.read_excel(file)
+    supplier_df["Capability"] = supplier_df["Capability"].fillna("")
+    texts = supplier_df["Capability"].tolist()
 
-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 embeddings
+    supplier_embeddings = model.encode(texts, convert_to_tensor=True, normalize_embeddings=True)
+
+    return f"{len(supplier_df)} suppliers loaded and processed."
+
+def view_supplier_data():
+    return supplier_df if supplier_df is not None else pd.DataFrame()
+
+def match_suppliers(event_description, top_k=5):
+    if supplier_embeddings is None:
+        return "Please upload and process supplier data first."
     
-    # Compute cosine similarity
+    event_embedding = model.encode(event_description, convert_to_tensor=True, normalize_embeddings=True).unsqueeze(0)
     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],
+            "Supplier Name": supplier_df.loc[idx, "Supplier Name"],
             "Match Score": round(scores[idx].item(), 4),
-            "Capabilities": supplier_texts[idx]
+            "Capabilities": supplier_df.loc[idx, "Capability"]
         })
-    
+
     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."
-)
+with gr.Blocks() as demo:
+    gr.Markdown("## Supplier Matching App with Full Pipeline")
+
+    with gr.Tab("1. Upload Supplier Data"):
+        file_input = gr.File(label="Upload Excel file", file_types=[".xlsx"])
+        load_button = gr.Button("Load and Process")
+        load_output = gr.Textbox(label="Status")
+        load_button.click(load_and_process_supplier_data, inputs=file_input, outputs=load_output)
+
+    with gr.Tab("2. View Supplier Data"):
+        view_button = gr.Button("Show Data")
+        data_output = gr.Dataframe()
+        view_button.click(view_supplier_data, outputs=data_output)
+
+    with gr.Tab("3. Match Use Case to Suppliers"):
+        input_text = gr.Textbox(lines=3, label="Describe Your Event/Use Case")
+        top_k_input = gr.Slider(minimum=1, maximum=10, value=5, label="Top K Matches")
+        match_button = gr.Button("Match Suppliers")
+        match_output = gr.Dataframe()
+        match_button.click(match_suppliers, inputs=[input_text, top_k_input], outputs=match_output)
 
 if __name__ == "__main__":
     demo.launch()