frst

.dockerignore

@@ -0,0 +1,20 @@
+__pycache__
+*.pyc
+*.pyo
+*.pyd
+.Python
+*.so
+*.egg
+*.egg-info
+dist
+build
+.git
+.gitignore
+.vscode
+.idea
+*.swp
+*.swo
+*~
+.DS_Store
+README.md
+.dockerignore

Dockerfile

@@ -0,0 +1,35 @@
+FROM python:3.10-slim
+
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    gcc \
+    g++ \
+    && rm -rf /var/lib/apt/lists/*
+
+# Upgrade pip first
+RUN pip install --upgrade pip
+
+# Copy requirements and install Python dependencies
+COPY requirements.txt .
+
+# Install packages separately for better caching and reliability
+RUN pip install --default-timeout=1000 --no-cache-dir flask flask-cors
+RUN pip install --default-timeout=1000 --no-cache-dir transformers==4.42.4
+RUN pip install --default-timeout=1000 --no-cache-dir torch==2.3.1
+
+# Copy application code
+COPY app.py .
+
+# Copy templates folder
+COPY templates/ ./templates/
+
+# Copy model files
+COPY models/ ./models/
+
+# Expose port
+EXPOSE 7860
+
+# Run the application
+CMD ["python", "app.py"]

README.md

@@ -1,10 +1,114 @@
 ---
 title: Basic Chatbot
-emoji: 🏆
+emoji: 💬
 colorFrom: blue
-colorTo: yellow
+colorTo: purple
 sdk: docker
+app_port: 7860
 pinned: false
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# AI Chatbot Flask App
+
+Flask service for a fine-tuned GPT-2 conversational chatbot model.
+
+## Setup
+
+### Local Development
+
+1. Install dependencies:
+```bash
+pip install -r requirements.txt
+```
+
+2. Run the app:
+```bash
+python app.py
+```
+
+3. Access the application:
+- **Web Interface**: http://localhost:7860 (Chat directly in your browser!)
+- API Info: http://localhost:7860/api
+- Health check: http://localhost:7860/health
+
+### Docker
+
+1. Build the image:
+```bash
+docker build -t chatbot-api .
+```
+
+2. Run the container:
+```bash
+docker run -p 7860:7860 chatbot-api
+```
+
+## API Endpoints
+
+### Web Interface
+Visit the root URL to access the interactive chat interface.
+
+### GET `/api`
+Health check and info
+```bash
+curl http://localhost:7860/api
+```
+
+### GET `/health`
+Detailed health status
+```bash
+curl http://localhost:7860/health
+```
+
+### POST `/chat`
+Generate chatbot response
+
+Request body:
+```json
+{
+  "user": ["Hello!", "How are you?"],
+  "ai": ["Hi there!"]
+}
+```
+
+Example:
+```bash
+curl -X POST http://localhost:7860/chat \
+  -H "Content-Type: application/json" \
+  -d '{"user": ["Hello!"], "ai": []}'
+```
+
+Response:
+```json
+{
+  "response": "Hi there! How can I help you today",
+  "device": "cuda:0"
+}
+```
+
+## Model
+
+- **Model**: Fine-tuned GPT-2
+- **Location**: `./models/fine-tuned-gpt2`
+- **Type**: Conversational AI Chatbot
+- **Port**: 7860
+
+## Features
+
+- CORS enabled for all origins
+- Automatic GPU detection and usage
+- Conversation history support (last 7 exchanges)
+- Clean chat interface
+- Real-time responses
+- Message history management
+- Typing indicators
+
+## Chat Interface
+
+The web interface provides:
+- Real-time chat with the AI
+- Conversation history
+- Typing indicators
+- Clear chat functionality
+- Responsive design for mobile and desktop
+

app.py

@@ -0,0 +1,112 @@
+from flask import Flask, jsonify, request, render_template
+from flask_cors import CORS
+from transformers import GPT2Tokenizer, GPT2LMHeadModel
+import torch
+
+app = Flask(__name__)
+CORS(app)
+
+# Global variables for model and tokenizer
+MODEL_PATH = "./models/fine-tuned-gpt2"
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+tokenizer = None
+model = None
+
+def load_chatbot_model():
+    """Load the chatbot model and tokenizer"""
+    global tokenizer, model
+    if model is None:
+        print(f"Loading chatbot model from {MODEL_PATH}...")
+        print(f"Using device: {device}")
+        
+        tokenizer = GPT2Tokenizer.from_pretrained(MODEL_PATH)
+        model = GPT2LMHeadModel.from_pretrained(MODEL_PATH)
+        model.to(device)
+        
+        print("Model loaded successfully!")
+
+# Load model on startup
+load_chatbot_model()
+
+@app.route("/")
+def index():
+    """Serve the chat interface"""
+    return render_template('index.html')
+
+@app.route("/api")
+def root():
+    return jsonify({
+        "message": "Chatbot API",
+        "status": "running",
+        "model": "fine-tuned-gpt2",
+        "device": str(device)
+    })
+
+@app.route("/health")
+def health():
+    return jsonify({
+        "status": "healthy",
+        "model_loaded": model is not None,
+        "device": str(device)
+    })
+
+@app.route("/chat", methods=["POST"])
+def chat():
+    """
+    Generate a chatbot response based on conversation history
+    """
+    if model is None or tokenizer is None:
+        return jsonify({"error": "Model not loaded"}), 500
+    
+    try:
+        data = request.get_json()
+        user_messages = data.get("user", [])
+        ai_messages = data.get("ai", [])
+        
+        # Build conversation history
+        combined_prompt = ""
+        
+        # Limit history to last 7 exchanges
+        user_msgs = user_messages[-7:] if len(user_messages) > 7 else user_messages
+        ai_msgs = ai_messages[-6:] if len(ai_messages) > 6 else ai_messages
+        
+        # Add conversation history
+        for user_message, ai_message in zip(user_msgs[:-1], ai_msgs):
+            combined_prompt += f"<user> {user_message}{tokenizer.eos_token}<AI> {ai_message}{tokenizer.eos_token}"
+        
+        # Add current message
+        if user_msgs:
+            combined_prompt += f"<user> {user_msgs[-1]}{tokenizer.eos_token}<AI>"
+        
+        # Tokenize and generate
+        inputs = tokenizer.encode(combined_prompt, return_tensors="pt").to(device)
+        attention_mask = torch.ones(inputs.shape, device=device)
+        
+        outputs = model.generate(
+            inputs,
+            max_new_tokens=50,
+            num_beams=5,
+            early_stopping=True,
+            no_repeat_ngram_size=2,
+            temperature=0.7,
+            top_k=50,
+            top_p=0.95,
+            pad_token_id=tokenizer.eos_token_id,
+            attention_mask=attention_mask,
+            repetition_penalty=1.2
+        )
+        
+        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+        # Extract only the new response
+        response = response.replace(combined_prompt, "").split(".")[0].strip()
+        
+        return jsonify({
+            "response": response,
+            "device": str(device)
+        })
+    
+    except Exception as e:
+        return jsonify({"error": str(e)}), 500
+
+if __name__ == "__main__":
+    app.run(host="0.0.0.0", port=7860, debug=False)

models/fine-tuned-gpt2/config.json

@@ -0,0 +1,39 @@
+{
+  "_name_or_path": "gpt2",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.0",
+  "use_cache": true,
+  "vocab_size": 50257
+}

models/fine-tuned-gpt2/config.json:Zone.Identifier

@@ -0,0 +1,3 @@
+[ZoneTransfer]
+ZoneId=3
+HostUrl=https://www.kaggle.com/

models/fine-tuned-gpt2/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "4.44.0"
+}

models/fine-tuned-gpt2/generation_config.json:Zone.Identifier

@@ -0,0 +1,3 @@
+[ZoneTransfer]
+ZoneId=3
+HostUrl=https://www.kaggle.com/

models/fine-tuned-gpt2/merges.txt:Zone.Identifier

@@ -0,0 +1,3 @@
+[ZoneTransfer]
+ZoneId=3
+HostUrl=https://www.kaggle.com/

models/fine-tuned-gpt2/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8be8247018b9ae965bcf6d6e3edaa797753fcf42623b65efa34973d31dae6aa3
+size 497774208

models/fine-tuned-gpt2/model.safetensors:Zone.Identifier

@@ -0,0 +1,3 @@
+[ZoneTransfer]
+ZoneId=3
+HostUrl=https://www.kaggle.com/

models/fine-tuned-gpt2/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

models/fine-tuned-gpt2/special_tokens_map.json:Zone.Identifier

@@ -0,0 +1,3 @@
+[ZoneTransfer]
+ZoneId=3
+HostUrl=https://www.kaggle.com/

models/fine-tuned-gpt2/tokenizer_config.json

@@ -0,0 +1,22 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 1024,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

models/fine-tuned-gpt2/tokenizer_config.json:Zone.Identifier

@@ -0,0 +1,3 @@
+[ZoneTransfer]
+ZoneId=3
+HostUrl=https://www.kaggle.com/

models/fine-tuned-gpt2/vocab.json:Zone.Identifier

@@ -0,0 +1,3 @@
+[ZoneTransfer]
+ZoneId=3
+HostUrl=https://www.kaggle.com/

requirements.txt

@@ -0,0 +1,4 @@
+flask
+flask-cors
+transformers==4.42.4
+torch==2.3.1

templates/index.html

@@ -0,0 +1,334 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>AI Chatbot - Chat with GPT-2</title>
+    <style>
+        * { margin: 0; padding: 0; box-sizing: border-box; }
+        
+        body {
+            font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, Oxygen, Ubuntu, Cantarell, sans-serif;
+            background: linear-gradient(135deg, #3b82f6 0%, #8b5cf6 100%);
+            min-height: 100vh;
+            padding: 20px;
+            display: flex;
+            flex-direction: column;
+        }
+        
+        .container {
+            max-width: 800px;
+            margin: 0 auto;
+            width: 100%;
+            display: flex;
+            flex-direction: column;
+            height: calc(100vh - 40px);
+        }
+        
+        header {
+            text-align: center;
+            color: white;
+            padding: 20px 0;
+        }
+        
+        h1 {
+            font-size: 2rem;
+            font-weight: 700;
+            text-shadow: 2px 2px 4px rgba(0,0,0,0.2);
+            margin-bottom: 5px;
+        }
+        
+        .subtitle {
+            opacity: 0.9;
+            font-size: 1rem;
+        }
+        
+        .chat-container {
+            flex: 1;
+            background: white;
+            border-radius: 16px;
+            box-shadow: 0 20px 60px rgba(0,0,0,0.3);
+            display: flex;
+            flex-direction: column;
+            overflow: hidden;
+        }
+        
+        .messages {
+            flex: 1;
+            overflow-y: auto;
+            padding: 20px;
+            display: flex;
+            flex-direction: column;
+            gap: 15px;
+        }
+        
+        .message {
+            max-width: 80%;
+            padding: 12px 16px;
+            border-radius: 12px;
+            word-wrap: break-word;
+            animation: slideIn 0.3s ease;
+        }
+        
+        @keyframes slideIn {
+            from { opacity: 0; transform: translateY(10px); }
+            to { opacity: 1; transform: translateY(0); }
+        }
+        
+        .message.user {
+            align-self: flex-end;
+            background: linear-gradient(135deg, #3b82f6 0%, #8b5cf6 100%);
+            color: white;
+            border-bottom-right-radius: 4px;
+        }
+        
+        .message.ai {
+            align-self: flex-start;
+            background: #f3f4f6;
+            color: #1f2937;
+            border-bottom-left-radius: 4px;
+        }
+        
+        .message.ai::before {
+            content: '🤖 ';
+        }
+        
+        .message.user::before {
+            content: '👤 ';
+        }
+        
+        .typing-indicator {
+            align-self: flex-start;
+            padding: 12px 16px;
+            background: #f3f4f6;
+            border-radius: 12px;
+            border-bottom-left-radius: 4px;
+            display: none;
+        }
+        
+        .typing-indicator.show {
+            display: block;
+        }
+        
+        .typing-indicator span {
+            height: 8px;
+            width: 8px;
+            background: #9ca3af;
+            border-radius: 50%;
+            display: inline-block;
+            margin: 0 2px;
+            animation: bounce 1.4s infinite ease-in-out;
+        }
+        
+        .typing-indicator span:nth-child(1) { animation-delay: -0.32s; }
+        .typing-indicator span:nth-child(2) { animation-delay: -0.16s; }
+        
+        @keyframes bounce {
+            0%, 80%, 100% { transform: scale(0); }
+            40% { transform: scale(1); }
+        }
+        
+        .input-area {
+            padding: 20px;
+            border-top: 1px solid #e5e7eb;
+            background: #fafafa;
+        }
+        
+        .input-container {
+            display: flex;
+            gap: 10px;
+        }
+        
+        #messageInput {
+            flex: 1;
+            padding: 12px 16px;
+            border: 2px solid #e5e7eb;
+            border-radius: 24px;
+            font-size: 1rem;
+            outline: none;
+            transition: border-color 0.2s;
+        }
+        
+        #messageInput:focus {
+            border-color: #3b82f6;
+        }
+        
+        #sendBtn {
+            padding: 12px 24px;
+            background: linear-gradient(135deg, #3b82f6 0%, #8b5cf6 100%);
+            color: white;
+            border: none;
+            border-radius: 24px;
+            font-weight: 600;
+            cursor: pointer;
+            transition: all 0.2s;
+        }
+        
+        #sendBtn:hover {
+            transform: translateY(-2px);
+            box-shadow: 0 5px 15px rgba(59, 130, 246, 0.3);
+        }
+        
+        #sendBtn:disabled {
+            opacity: 0.5;
+            cursor: not-allowed;
+            transform: none;
+        }
+        
+        .clear-btn {
+            text-align: center;
+            padding: 10px;
+            border-top: 1px solid #e5e7eb;
+            background: #fafafa;
+        }
+        
+        .clear-btn button {
+            padding: 8px 16px;
+            background: transparent;
+            color: #6b7280;
+            border: 1px solid #d1d5db;
+            border-radius: 8px;
+            font-size: 0.875rem;
+            cursor: pointer;
+            transition: all 0.2s;
+        }
+        
+        .clear-btn button:hover {
+            background: #f3f4f6;
+            color: #374151;
+        }
+        
+        .empty-state {
+            text-align: center;
+            color: #9ca3af;
+            padding: 40px 20px;
+        }
+        
+        .empty-state h3 {
+            font-size: 1.5rem;
+            margin-bottom: 10px;
+        }
+        
+        @media (max-width: 640px) {
+            h1 { font-size: 1.5rem; }
+            .message { max-width: 90%; }
+            #sendBtn { padding: 12px 20px; }
+        }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <header>
+            <h1>💬 AI Chatbot</h1>
+            <p class="subtitle">Chat with fine-tuned GPT-2</p>
+        </header>
+        
+        <div class="chat-container">
+            <div class="messages" id="messages">
+                <div class="empty-state">
+                    <h3>👋 Hello!</h3>
+                    <p>Start a conversation by typing a message below</p>
+                </div>
+            </div>
+            
+            <div class="clear-btn">
+                <button id="clearBtn">Clear Chat</button>
+            </div>
+            
+            <div class="input-area">
+                <div class="input-container">
+                    <input type="text" id="messageInput" placeholder="Type your message..." autocomplete="off">
+                    <button id="sendBtn">Send</button>
+                </div>
+            </div>
+        </div>
+    </div>
+    
+    <script>
+        const messagesDiv = document.getElementById('messages');
+        const messageInput = document.getElementById('messageInput');
+        const sendBtn = document.getElementById('sendBtn');
+        const clearBtn = document.getElementById('clearBtn');
+        
+        let conversationHistory = { user: [], ai: [] };
+        
+        function addMessage(text, isUser) {
+            const emptyState = messagesDiv.querySelector('.empty-state');
+            if (emptyState) emptyState.remove();
+            
+            const messageDiv = document.createElement('div');
+            messageDiv.className = `message ${isUser ? 'user' : 'ai'}`;
+            messageDiv.textContent = text;
+            messagesDiv.appendChild(messageDiv);
+            messagesDiv.scrollTop = messagesDiv.scrollHeight;
+        }
+        
+        function showTyping() {
+            const typing = document.createElement('div');
+            typing.className = 'typing-indicator show';
+            typing.id = 'typing';
+            typing.innerHTML = '<span></span><span></span><span></span>';
+            messagesDiv.appendChild(typing);
+            messagesDiv.scrollTop = messagesDiv.scrollHeight;
+        }
+        
+        function hideTyping() {
+            const typing = document.getElementById('typing');
+            if (typing) typing.remove();
+        }
+        
+        async function sendMessage() {
+            const message = messageInput.value.trim();
+            if (!message) return;
+            
+            addMessage(message, true);
+            conversationHistory.user.push(message);
+            messageInput.value = '';
+            sendBtn.disabled = true;
+            messageInput.disabled = true;
+            
+            showTyping();
+            
+            try {
+                const response = await fetch('/chat', {
+                    method: 'POST',
+                    headers: { 'Content-Type': 'application/json' },
+                    body: JSON.stringify(conversationHistory)
+                });
+                
+                if (!response.ok) throw new Error('Failed to get response');
+                
+                const data = await response.json();
+                hideTyping();
+                
+                const aiResponse = data.response || 'Sorry, I could not generate a response.';
+                addMessage(aiResponse, false);
+                conversationHistory.ai.push(aiResponse);
+                
+            } catch (error) {
+                hideTyping();
+                addMessage('Sorry, something went wrong. Please try again.', false);
+                conversationHistory.user.pop(); // Remove last user message on error
+            } finally {
+                sendBtn.disabled = false;
+                messageInput.disabled = false;
+                messageInput.focus();
+            }
+        }
+        
+        sendBtn.addEventListener('click', sendMessage);
+        messageInput.addEventListener('keypress', (e) => {
+            if (e.key === 'Enter') sendMessage();
+        });
+        
+        clearBtn.addEventListener('click', () => {
+            messagesDiv.innerHTML = '<div class="empty-state"><h3>👋 Hello!</h3><p>Start a conversation by typing a message below</p></div>';
+            conversationHistory = { user: [], ai: [] };
+            messageInput.value = '';
+            messageInput.focus();
+        });
+        
+        messageInput.focus();
+    </script>
+</body>
+</html>