Update app.py

app.py

@@ -1,82 +1,76 @@
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
+from peft import PeftModel, PeftConfig
 import gradio as gr
-from huggingface_hub import InferenceClient
 
+# === Load your LoRA correctly ===
+peft_model_id = "rishu834763/java-explainer-lora"
 
-from transformers import AutoModelForCausalLM
-from peft import PeftModel
+config = PeftConfig.from_pretrained(peft_model_id)
+base_model_name = config.base_model_name_or_path  # this will be mistralai/Mistral-7B-Instruct-v0.2
 
-# This automatically loads the base model mentioned in adapter_config.json
+# Load base model (with quantization if you want to fit in free tier)
 model = AutoModelForCausalLM.from_pretrained(
-    "rishu834763/java-explainer-lora",
+    base_model_name,
+    torch_dtype=torch.bfloat16,
     device_map="auto",
-    torch_dtype="auto",
-    trust_remote_code=True  # only needed for some gated/gated models
+    # Remove the two lines below if you have enough VRAM or a paid Space
+    # load_in_4bit=True,
+    # quantization_config=BitsAndBytesConfig(
+    #     load_in_4bit=True,
+    #     bnb_4bit_compute_dtype=torch.bfloat16,
+    #     bnb_4bit_use_double_quant=True,
+    #     bnb_4bit_quant_type="nf4"
+    # ),
 )
-model = PeftModel.from_pretrained(model, "rishu834763/java-explainer-lora")
 
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
-    """
-    For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-    """
-    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
+model = PeftModel.from_pretrained(model, peft_model_id)
+# Optional but recommended: merge so inference is faster and uses less VRAM
+model = model.merge_and_unload()
 
-    messages = [{"role": "system", "content": system_message}]
+tokenizer = AutoTokenizer.from_pretrained(base_model_name)
+if tokenizer.pad_token is None:
+    tokenizer.pad_token = tokenizer.eos_token
 
-    messages.extend(history)
+# Create the pipeline using YOUR model and tokenizer
+pipe = pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+)
 
+# === This is the most important part ===
+def chat(message, history):
+    messages = []
+    for user_msg, assistant_msg in history:
+        messages.append({"role": "user", "content": user_msg})
+        if assistant_msg:
+            messages.append({"role": "assistant", "content": assistant_msg})
     messages.append({"role": "user", "content": message})
 
-    response = ""
-
-    for message in client.chat_completion(
+    outputs = pipe(
         messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
-
-        response += token
-        yield response
+        max_new_tokens=512,
+        do_sample=True,
+        temperature=0.7,
+        top_p=0.9,
+        pad_token_id=tokenizer.eos_token_id
+    )
+    response = outputs[0]["generated_text"][-1]["content"]
+    return response
 
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-chatbot = gr.ChatInterface(
-    respond,
-    type="messages",
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
+# === Build the Gradio interface ===
+demo = gr.ChatInterface(
+    fn=chat,
+    title="Java Explainer (Mistral-7B + your LoRA)",
+    description="Ask anything about Java code → I will explain it using your fine-tuned model",
+    examples=[
+        "Explain this Java code: public class HelloWorld { public static void main(String[] args) { System.out.println(\"Hello, World!\"); } }",
+        "What does synchronized keyword do in Java?"
     ],
+    cache_examples=False,
 )
 
-with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
-    chatbot.render()
-
-
-if __name__ == "__main__":
-    demo.launch()
+demo.launch()