Create app.py

app.py

@@ -0,0 +1,46 @@
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
+from peft import PeftModel, PeftConfig
+import gradio as gr
+
+# Your LoRA
+PEFT_ID = "rishu834763/java-explainer-lora"
+
+config = PeftConfig.from_pretrained(PEFT_ID)
+base = config.base_model_name_or_path
+
+model = AutoModelForCausalLM.from_pretrained(
+    base,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+    load_in_4bit=True
+)
+model = PeftModel.from_pretrained(model, PEFT_ID)
+model = model.merge_and_unload()
+
+tokenizer = AutoTokenizer.from_pretrained(base)
+if tokenizer.pad_token is None:
+    tokenizer.pad_token = tokenizer.eos_token
+
+pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, max_new_tokens=1024, temperature=0.6, do_sample=True)
+
+def chat(message, history):
+    msgs = []
+    for u, a in history:
+        msgs.append({"role": "user", "content": u})
+        if a: msgs.append({"role": "assistant", "content": a})
+    msgs.append({"role": "user", "content": message})
+    
+    out = pipe(msgs)
+    return out[0]["generated_text"][-1]["content"]
+
+gr.ChatInterface(
+    chat,
+    title="Java Explainer – Your Own Model (No OpenAI)",
+    description="This is 100% your fine-tuned Java LoRA running locally on Hugging Face",
+    examples=[
+        "Explain this Java code: public static void main(String[] args) { System.out.println(\"Hello\"); }",
+        "What does @Override do in Java?",
+        "Difference between HashMap and Hashtable?"
+    ]
+).queue().launch()