Delete app.py

app.py

@@ -1,129 +0,0 @@
-# app.py — Gilbert Multitask AI (LoRA)
-import gradio as gr
-import torch
-from transformers import AutoModelForCausalLM, AutoTokenizer
-from peft import PeftModel
-
-# Model config
-MODEL_NAME = "GilbertAkham/gilbert-qwen-multitask-lora"
-BASE_MODEL = "Qwen/Qwen1.5-1.8B-Chat"
-
-
-# ------------------------
-# MODEL LOADING
-# ------------------------
-class MultitaskInference:
-    def __init__(self):
-        self.model = None
-        self.tokenizer = None
-        self.device = "cuda" if torch.cuda.is_available() else "cpu"
-        self.load_model()
-
-    def load_model(self):
-        """Load base + LoRA model"""
-        print("🔄 Loading tokenizer...")
-        self.tokenizer = AutoTokenizer.from_pretrained(BASE_MODEL, trust_remote_code=True)
-
-        print("🔄 Loading base model...")
-        base_model = AutoModelForCausalLM.from_pretrained(
-            BASE_MODEL,
-            torch_dtype=torch.float16 if self.device == "cuda" else torch.float32,
-            device_map="auto" if self.device == "cuda" else None,
-            trust_remote_code=True,
-        )
-
-        print("🔄 Loading LoRA adapter...")
-        self.model = PeftModel.from_pretrained(base_model, MODEL_NAME)
-        self.model.to(self.device)
-        self.model.eval()
-        print("✅ Model loaded successfully!")
-
-    def generate(self, task_type, text, max_tokens=512, temperature=0.7, top_p=0.9):
-        """Generate multitask response"""
-        task_prompts = {
-            "email": "Draft an email reply",
-            "story": "Continue the story",
-            "tech": "Answer the technical question",
-            "summary": "Summarize the content",
-            "chat": "Provide a helpful chat response"
-        }
-
-        prompt = f"### Task: {task_prompts[task_type]}\n\n### Input:\n{text}\n\n### Output:\n"
-        try:
-            inputs = self.tokenizer(prompt, return_tensors="pt", truncation=True, max_length=1024).to(self.device)
-
-            with torch.no_grad():
-                outputs = self.model.generate(
-                    **inputs,
-                    max_new_tokens=max_tokens,
-                    temperature=temperature,
-                    do_sample=True,
-                    top_p=top_p,
-                    repetition_penalty=1.1,
-                    pad_token_id=self.tokenizer.eos_token_id,
-                )
-
-            response = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
-            if "### Output:" in response:
-                response = response.split("### Output:")[-1].strip()
-            return response
-
-        except Exception as e:
-            return f"❌ Error generating response: {e}"
-
-
-# ------------------------
-# GRADIO INTERFACE
-# ------------------------
-engine = MultitaskInference()
-
-
-def chat_response(message, history, task_type, max_tokens, temperature, top_p):
-    """Chat handler for Gradio ChatInterface"""
-    try:
-        reply = engine.generate(
-            task_type=task_type,
-            text=message,
-            max_tokens=max_tokens,
-            temperature=temperature,
-            top_p=top_p
-        )
-        yield reply
-    except Exception as e:
-        yield f"❌ Error: {e}"
-
-
-# ------------------------
-# BUILD CHAT APP
-# ------------------------
-chatbot = gr.ChatInterface(
-    fn=chat_response,
-    type="messages",
-    additional_inputs=[
-        gr.Dropdown(
-            choices=["chat", "email", "story", "tech", "summary"],
-            value="chat",
-            label="🎯 Task Type",
-            info="Select the text generation mode",
-        ),
-        gr.Slider(minimum=64, maximum=1024, value=512, step=32, label="📏 Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=1.0, value=0.7, step=0.05, label="🌡️ Temperature"),
-        gr.Slider(minimum=0.1, maximum=1.0, value=0.9, step=0.05, label="🎲 Top-p"),
-    ],
-    title="🚀 Gilbert Multitask AI",
-    description=(
-        "**Base Model:** Qwen1.5-1.8B-Chat\n\n"
-        "LoRA fine-tuned for: Email drafting, story continuation, tech Q&A, summarization, and chat responses."
-    ),
-    theme=gr.themes.Soft(),
-    examples=[
-        ["Write a professional email update for a client about completing a milestone."],
-        ["Continue the story: The spaceship hummed as Captain Lira adjusted the controls..."],
-        ["How can I fix a 'ModuleNotFoundError' in Python?"],
-        ["Summarize: Artificial intelligence is transforming industries through automation and insights."],
-        ["Hey, I can’t access the company VPN. What should I do?"],
-    ],
-)
-
-if __name__ == "__main__":
-    chatbot.launch(server_name="0.0.0.0", server_port=7860, share=True)