Upload app.py

app.py

@@ -101,19 +101,28 @@ def load_model():
 
 model = load_model()
 
-# Define encode and decode functions
-chars = sorted(list(set("your_training_text_here")))  # Replace with the character set used in training
+# Define a comprehensive character set based on training data
+chars = sorted(list(set("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789 .,!?-:;'\"\n")))
 stoi = {ch: i for i, ch in enumerate(chars)}
 itos = {i: ch for i, ch in enumerate(chars)}
-encode = lambda s: [stoi[c] for c in s]
+encode = lambda s: [stoi[c] for c in s if c in stoi]  # Ensures only known characters are encoded
 decode = lambda l: ''.join([itos[i] for i in l])
 
 # Function to generate text using the model
 def generate_text(prompt):
-    context = torch.tensor([encode(prompt)], dtype=torch.long)
-    with torch.no_grad():
-        generated = model.generate(context, max_new_tokens=250)  # Adjust as needed
-    return decode(generated[0].tolist())
+    try:
+        print(f"Received prompt: {prompt}")
+        context = torch.tensor([encode(prompt)], dtype=torch.long)
+        print(f"Encoded prompt: {context}")
+        with torch.no_grad():
+            generated = model.generate(context, max_new_tokens=250)  # Adjust as needed
+            print(f"Generated tensor: {generated}")
+        result = decode(generated[0].tolist())
+        print(f"Decoded result: {result}")
+        return result
+    except Exception as e:
+        print(f"Error during generation: {e}")
+        return f"Error: {str(e)}"
 
 # Create a Gradio interface
 interface = gr.Interface(
@@ -125,4 +134,4 @@ interface = gr.Interface(
 )
 
 # Launch the interface
-interface.launch()
+interface.launch(share=True)