Update app.py

app.py

@@ -1,64 +1,84 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+import torch
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
 
-"""
-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("HuggingFaceH4/zephyr-7b-beta")
+# Lightweight model for fast inference
+model_name = "Salesforce/codet5p-220m"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSeq2SeqLM.from_pretrained(
+    model_name,
+    torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32
+)
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
 
+# Prompt templates
+language_prompts = {
+    "Python": "Fix this Python code:\n",
+    "C": "Fix this C code:\n",
+    "C++": "Fix this C++ code:\n",
+    "JavaScript": "Fix this JavaScript code:\n"
+}
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# Debugger logic
+def eternos_debugger(code, error, language):
+    if not code.strip():
+        return "❌ Please provide code."
+    prompt = f"{language_prompts[language]}{code}\nError:\n{error}\nCorrected code:\n"
+    inputs = tokenizer(prompt, return_tensors="pt", truncation=True, max_length=512).to(device)
+    outputs = model.generate(
+        **inputs,
+        max_new_tokens=128,
+        temperature=0.1,
+        do_sample=False,
+        pad_token_id=tokenizer.pad_token_id or tokenizer.eos_token_id
+    )
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return response.strip()
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+# Reinforcement learning simulation
+def rl_simulation():
+    import gym
+    import numpy as np
 
-    messages.append({"role": "user", "content": message})
+    env = gym.make("FrozenLake-v1", is_slippery=False)
+    Q = np.zeros([env.observation_space.n, env.action_space.n])
+    episodes = 500
+    learning_rate = 0.8
+    discount_factor = 0.95
 
-    response = ""
+    for ep in range(episodes):
+        state = env.reset()[0]
+        done = False
+        while not done:
+            action = np.argmax(Q[state, :] + np.random.randn(1, env.action_space.n) * (1.0 / (ep + 1)))
+            new_state, reward, done, _, _ = env.step(action)
+            Q[state, action] += learning_rate * (reward + discount_factor * np.max(Q[new_state, :]) - Q[state, action])
+            state = new_state
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+    return "🧠 RL training complete! Agent learned to navigate FrozenLake."
 
-        response += token
-        yield response
+# UI Layout (no background CSS for white theme)
+with gr.Blocks(theme=gr.themes.Soft()) as demo:
+    with gr.Tab("Eternos Debugger"):
+        gr.Markdown("## ⚙️ Eternos — AI Code Debugger")
+        gr.Markdown("Supports Python, C, C++, JavaScript — powered by CodeT5p (Fast Edition)")
 
+        with gr.Row():
+            code_input = gr.Textbox(label="📝 Your Code", lines=12)
+            error_input = gr.Textbox(label="⚠️ Error Message (optional)", lines=4)
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    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)",
-        ),
-    ],
-)
+        language_input = gr.Dropdown(["Python", "C", "C++", "JavaScript"], label="🌐 Language", value="Python")
+        output_code = gr.Code(label="✅ Suggested Fix")
+        run_btn = gr.Button("🛠️ Fix Code")
+
+        run_btn.click(fn=eternos_debugger, inputs=[code_input, error_input, language_input], outputs=output_code)
 
+    with gr.Tab("Reinforcement Learning Demo"):
+        gr.Markdown("## 🤖 Reinforcement Learning Example")
+        gr.Markdown("Simulates a Q-learning agent on FrozenLake.")
+        rl_output = gr.Textbox(label="🔄 Output", lines=4)
+        rl_btn = gr.Button("🏁 Run RL Simulation")
+        rl_btn.click(fn=rl_simulation, inputs=[], outputs=rl_output)
 
-if __name__ == "__main__":
-    demo.launch()
+demo.launch()