Update app.py

app.py

@@ -1,42 +1,38 @@
-import warnings
 import gradio as gr
-from proxy_model import RemoteModelProxy
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
 
-# Suppress the FutureWarning
-warnings.filterwarnings("ignore", category=FutureWarning, module="torch")
+# Model and Tokenizer from Hugging Face Hub
+model_name = "deepseek-ai/DeepSeek-V3"
 
-# Load the model via the proxy
-model_proxy = RemoteModelProxy("deepseek-ai/DeepSeek-V3")
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype=torch.bfloat16,  # Use bfloat16 for faster and less memory-intensive inference if possible
+    trust_remote_code=True,      # Important for models with custom code
+    device_map="auto"           # Automatically use available GPU if possible
+)
 
-# Define the text classification function
-def classify_text(text):
-    try:
-        result = model_proxy.classify_text(text)
-        return result
-    except Exception as e:
-        print(f"Error during text classification: {e}")
-        return {
-            "Predicted Class": "Error",
-            "Probabilities": []
-        }
+def generate_response(prompt, history=[]):
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    outputs = model.generate(**inputs, max_new_tokens=500) # Adjust max_new_tokens as needed
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
 
-# Create a Gradio interface
-try:
-    iface = gr.Interface(
-        fn=classify_text,  # Function to call
-        inputs=gr.Textbox(lines=2, placeholder="Enter text here..."),  # Input component
-        outputs=[
-            gr.Label(label="Predicted Class"),  # Output component for predicted class
-            gr.Label(label="Probabilities")  # Output component for probabilities
-        ],
-        title="DeepSeek-V3 Text Classification",
-        description="Classify text using the DeepSeek-V3 model."
-    )
-except Exception as e:
-    print(f"Failed to create Gradio interface: {e}")
+    # Basic chat history handling (optional, can be improved)
+    history.append((prompt, response)) # Append user prompt and model response to history
+    return response
 
-# Launch the interface
-try:
-    iface.launch()
-except Exception as e:
-    print(f"Failed to launch Gradio interface: {e}")
+iface = gr.ChatInterface(
+    fn=generate_response,
+    inputs=gr.Chatbox(lines=7, placeholder="Type your message here..."),
+    outputs="text",
+    title="DeepSeek-V3 Chatbot",
+    description="Chat with the DeepSeek-V3 model. Please be patient, initial loading might take a few minutes. For better performance, use a Space with a GPU.",
+    examples=[
+        "Hello, how are you?",
+        "What is the capital of France?",
+        "Tell me a joke."
+    ]
+)
+
+iface.launch(share=False)