.

app.py

@@ -1,70 +1,133 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
 
+# Load Phi-2 model and tokenizer
+model_name = "microsoft/phi-2"
+tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(model_name, trust_remote_code=True, torch_dtype=torch.float32)
 
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
+def generate_code(prompt, max_length=100, temperature=0.7, num_outputs=1):
     """
-    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
+    Generate code completion using Phi-2.
+    
+    Args:
+        prompt: Code prompt/prefix
+        max_length: Maximum length of generated code
+        temperature: Sampling temperature (higher = more creative)
+        num_outputs: Number of different completions to generate
+    
+    Returns:
+        JSON object with generated code
     """
-    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
-
-    messages = [{"role": "system", "content": system_message}]
-
-    messages.extend(history)
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        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
-
-
-"""
-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)",
-        ),
-    ],
-)
-
-with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
-    chatbot.render()
+    try:
+        # Tokenize input
+        inputs = tokenizer(prompt, return_tensors="pt", return_attention_mask=True)
+        
+        # Generate code
+        with torch.no_grad():
+            outputs = model.generate(
+                inputs["input_ids"],
+                attention_mask=inputs["attention_mask"],
+                max_length=max_length,
+                temperature=temperature,
+                num_return_sequences=num_outputs,
+                do_sample=True,
+                top_p=0.95,
+                pad_token_id=tokenizer.eos_token_id
+            )
+        
+        # Decode generated sequences
+        completions = []
+        for idx, output in enumerate(outputs):
+            generated_text = tokenizer.decode(output, skip_special_tokens=True)
+            completions.append({
+                "rank": idx + 1,
+                "generated_code": generated_text,
+                "continuation": generated_text[len(prompt):]
+            })
+        
+        return {
+            "prompt": prompt,
+            "completions": completions
+        }
+    
+    except Exception as e:
+        return {
+            "error": str(e),
+            "completions": []
+        }
 
+# Create Gradio interface
+with gr.Blocks(title="Phi-2 Code Generator") as demo:
+    gr.Markdown(
+        """
+        # Phi-2 Code Generator (2.7B)
+        
+        This model generates code completions using Microsoft's Phi-2 language model.
+        Enter a code prompt and the model will continue writing the code.
+        
+        ### Examples:
+        - `def add(x, y):`
+        - `import numpy as np\n# Calculate`
+        - `class Calculator:\n    def __init__(self):`
+        - `# Function to sort a list\ndef`
+        """
+    )
+    
+    with gr.Row():
+        with gr.Column():
+            code_input = gr.Textbox(
+                label="Code Prompt",
+                placeholder="Enter your code prompt...",
+                lines=5,
+                value="def fibonacci(n):"
+            )
+            max_length_slider = gr.Slider(
+                minimum=50,
+                maximum=500,
+                value=100,
+                step=10,
+                label="Max Length"
+            )
+            temperature_slider = gr.Slider(
+                minimum=0.1,
+                maximum=1.5,
+                value=0.7,
+                step=0.1,
+                label="Temperature (creativity)"
+            )
+            num_outputs_slider = gr.Slider(
+                minimum=1,
+                maximum=3,
+                value=1,
+                step=1,
+                label="Number of outputs"
+            )
+            generate_btn = gr.Button("Generate", variant="primary")
+        
+        with gr.Column():
+            output = gr.JSON(
+                label="Generated Code"
+            )
+    
+    # Examples
+    gr.Examples(
+        examples=[
+            ["def fibonacci(n):", 100, 0.7, 1],
+            ["import pandas as pd\n# Load and analyze data\n", 150, 0.7, 1],
+            ["class BinaryTree:\n    def __init__(self):", 120, 0.7, 1],
+            ["# Function to reverse a string\ndef reverse_string(s):", 100, 0.7, 1],
+            ["for i in range(10):", 80, 0.7, 1],
+        ],
+        inputs=[code_input, max_length_slider, temperature_slider, num_outputs_slider],
+    )
+    
+    generate_btn.click(
+        fn=generate_code,
+        inputs=[code_input, max_length_slider, temperature_slider, num_outputs_slider],
+        outputs=output
+    )
 
 if __name__ == "__main__":
     demo.launch()

requirements.txt

@@ -0,0 +1,3 @@
+transformers
+torch
+gradio