add the fine tuned BERT model with FAST API integrated in the Flask app

Dockerfile

@@ -1,20 +1,22 @@
-FROM python:3.9-slim
-
-WORKDIR /code
-
-# Install system dependencies
-RUN apt-get update && apt-get install -y git && rm -rf /var/lib/apt/lists/*
-
-# Copy requirements and install
-COPY requirements.txt .
-RUN pip install --no-cache-dir -r requirements.txt
-
-# Copy the rest of the code
-COPY . .
-
-# Expose FastAPI port
-EXPOSE 8000
-
-# Hugging Face Spaces expects the app to run on 0.0.0.0:8000
-ENV FLASK_APP=src.app
-CMD ["flask", "run", "--host=0.0.0.0", "--port=5000", "--no-debugger", "--no-reload"]
+FROM python:3.9-slim
+
+WORKDIR /app
+
+# Copy requirements first for better caching
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy the entire project
+COPY . .
+
+# Create necessary directories if they don't exist
+RUN mkdir -p /app/model /app/intent_classifier_model /app/intent_classifier_tokenizer
+
+# Expose the port
+EXPOSE 7860
+
+# Set environment variables
+ENV PYTHONPATH=/app
+
+# Run the FastAPI application
+CMD ["uvicorn", "model.api.api:app", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -1,10 +1,11 @@
 ---
 title: Intent Classifier Chatbot
 emoji: 🤖
-colorFrom: green
+colorFrom: blue
 colorTo: purple
 sdk: docker
 pinned: false
+app_port: 7860
 license: apache-2.0
 short_description: Intent Detection API using BERT and Flask
 ---
@@ -68,6 +69,50 @@ The project uses the **CLINC150 dataset**, a benchmark dataset for intent classi
 
 ---
 
+# Intent Classifier Chatbot
+
+A sophisticated intent classification system built with BERT and FastAPI that can predict user intents from natural language text.
+
+## Features
+
+- **Advanced NLP**: Uses BERT-based transformer model for accurate intent classification
+- **150+ Intent Classes**: Trained on the CLINC150 dataset with comprehensive intent coverage
+- **Real-time Prediction**: FastAPI backend for fast inference
+- **Clean UI**: Simple and intuitive web interface
+- **Production Ready**: Dockerized for easy deployment
+
+## How to Use
+
+1. Enter your message in the text area
+2. Click "Predict Intent" 
+3. See the AI's prediction of your intent
+
+Try examples like:
+- "Set an alarm for 7am" → Alarm
+- "Transfer money to John" → Transfer
+- "What's the weather like?" → Weather
+- "Book a flight to Paris" → Book Flight
+
+## Model Details
+
+- **Architecture**: BERT for Sequence Classification
+- **Dataset**: CLINC150 (151 intent classes including out-of-scope)
+- **Accuracy**: High performance on intent classification tasks
+- **Preprocessing**: Advanced tokenization and text normalization
+
+## Tech Stack
+
+- **Backend**: FastAPI, PyTorch, Transformers
+- **Frontend**: HTML, CSS, JavaScript
+- **Model**: BERT-base fine-tuned on CLINC150
+- **Deployment**: Docker, Hugging Face Spaces
+
+## Author
+
+**Saher Muhamed**
+- GitHub: [@sahermuhamed1](https://github.com/sahermuhamed1)
+- Email: sahermuhamed176@gmail.com
+
 ## 🤗 Hugging Face Spaces Configuration
 
 To deploy this project on [Hugging Face Spaces](https://huggingface.co/spaces), you can use a `README.md` and a `config.json` file to configure your Space for inference.  
@@ -87,6 +132,4 @@ Example `config.json` for inference API:
 
 - Make sure your `requirements.txt` lists all dependencies.
 - The `entrypoint` should point to your main app file (e.g., `app.py` or `main.py`).
-- For more details and advanced configuration, see the [Spaces config reference](https://huggingface.co/docs/hub/spaces-config-reference).
-
----
+- For more details and advanced configuration, see the [Spaces config reference](https://huggingface.co/docs/hub/spaces-config-reference).

model/api/api.py

@@ -1,10 +1,12 @@
-from fastapi import FastAPI
+from fastapi import FastAPI, Request
+from fastapi.responses import HTMLResponse
+from fastapi.staticfiles import StaticFiles
 from pydantic import BaseModel
 from transformers import BertForSequenceClassification, BertTokenizer
 import torch
 import os
 
-app = FastAPI()
+app = FastAPI(title="Intent Classifier API", description="BERT-based intent classification system")
 
 # Get the absolute path to the model directory
 BASE_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
@@ -13,7 +15,7 @@ BASE_DIR = os.path.dirname(BASE_DIR)
 MODEL_DIR = os.path.join(BASE_DIR, "intent_classifier_model")
 TOKENIZER_DIR = os.path.join(BASE_DIR, "intent_classifier_tokenizer")
 
-# Ensure model and tokenizer directories exist
+# Ensure model and tokenizer directories exist  
 if not os.path.isdir(MODEL_DIR):
     raise FileNotFoundError(f"Model directory not found: {MODEL_DIR}")
 if not os.path.isdir(TOKENIZER_DIR):
@@ -23,23 +25,355 @@ if not os.path.isdir(TOKENIZER_DIR):
 model = BertForSequenceClassification.from_pretrained(MODEL_DIR, local_files_only=True)
 tokenizer = BertTokenizer.from_pretrained(TOKENIZER_DIR, local_files_only=True)
 
-# Load intent label mapping
-from datasets import load_dataset
-dataset = load_dataset("clinc_oos", "small")
-int2str = dataset["train"].features["intent"].int2str
+# Complete CLINC150 intent labels in exact order (151 total)
+INTENT_LABELS = ['restaurant_reviews',
+ 'nutrition_info',
+ 'account_blocked',
+ 'oil_change_how',
+ 'time',
+ 'weather',
+ 'redeem_rewards',
+ 'interest_rate',
+ 'gas_type',
+ 'accept_reservations',
+ 'smart_home',
+ 'user_name',
+ 'report_lost_card',
+ 'repeat',
+ 'whisper_mode',
+ 'what_are_your_hobbies',
+ 'order',
+ 'jump_start',
+ 'schedule_meeting',
+ 'meeting_schedule',
+ 'freeze_account',
+ 'what_song',
+ 'meaning_of_life',
+ 'restaurant_reservation',
+ 'traffic',
+ 'make_call',
+ 'text',
+ 'bill_balance',
+ 'improve_credit_score',
+ 'change_language',
+ 'no',
+ 'measurement_conversion',
+ 'timer',
+ 'flip_coin',
+ 'do_you_have_pets',
+ 'balance',
+ 'tell_joke',
+ 'last_maintenance',
+ 'exchange_rate',
+ 'uber',
+ 'car_rental',
+ 'credit_limit',
+ 'oos',
+ 'shopping_list',
+ 'expiration_date',
+ 'routing',
+ 'meal_suggestion',
+ 'tire_change',
+ 'todo_list',
+ 'card_declined',
+ 'rewards_balance',
+ 'change_accent',
+ 'vaccines',
+ 'reminder_update',
+ 'food_last',
+ 'change_ai_name',
+ 'bill_due',
+ 'who_do_you_work_for',
+ 'share_location',
+ 'international_visa',
+ 'calendar',
+ 'translate',
+ 'carry_on',
+ 'book_flight',
+ 'insurance_change',
+ 'todo_list_update',
+ 'timezone',
+ 'cancel_reservation',
+ 'transactions',
+ 'credit_score',
+ 'report_fraud',
+ 'spending_history',
+ 'directions',
+ 'spelling',
+ 'insurance',
+ 'what_is_your_name',
+ 'reminder',
+ 'where_are_you_from',
+ 'distance',
+ 'payday',
+ 'flight_status',
+ 'find_phone',
+ 'greeting',
+ 'alarm',
+ 'order_status',
+ 'confirm_reservation',
+ 'cook_time',
+ 'damaged_card',
+ 'reset_settings',
+ 'pin_change',
+ 'replacement_card_duration',
+ 'new_card',
+ 'roll_dice',
+ 'income',
+ 'taxes',
+ 'date',
+ 'who_made_you',
+ 'pto_request',
+ 'tire_pressure',
+ 'how_old_are_you',
+ 'rollover_401k',
+ 'pto_request_status',
+ 'how_busy',
+ 'application_status',
+ 'recipe',
+ 'calendar_update',
+ 'play_music',
+ 'yes',
+ 'direct_deposit',
+ 'credit_limit_change',
+ 'gas',
+ 'pay_bill',
+ 'ingredients_list',
+ 'lost_luggage',
+ 'goodbye',
+ 'what_can_i_ask_you',
+ 'book_hotel',
+ 'are_you_a_bot',
+ 'next_song',
+ 'change_speed',
+ 'plug_type',
+ 'maybe',
+ 'w2',
+ 'oil_change_when',
+ 'thank_you',
+ 'shopping_list_update',
+ 'pto_balance',
+ 'order_checks',
+ 'travel_alert',
+ 'fun_fact',
+ 'sync_device',
+ 'schedule_maintenance',
+ 'apr',
+ 'transfer',
+ 'ingredient_substitution',
+ 'calories',
+ 'current_location',
+ 'international_fees',
+ 'calculator',
+ 'definition',
+ 'next_holiday',
+ 'update_playlist',
+ 'mpg',
+ 'min_payment',
+ 'change_user_name',
+ 'restaurant_suggestion',
+ 'travel_notification',
+ 'cancel',
+ 'pto_used',
+ 'travel_suggestion',
+ 'change_volume']
+def int2str(idx):
+    return INTENT_LABELS[idx] if 0 <= idx < len(INTENT_LABELS) else "unknown"
 
 class Query(BaseModel):
-    text: str
+    text: str = None
+    message: str = None
 
+# Add compatibility endpoint for both 'message' and 'text' fields
 @app.post("/predict")
-def predict_intent(query: Query):
-    inputs = tokenizer(query.text, return_tensors="pt", truncation=True, padding=True, max_length=128)
-    with torch.no_grad():
-        outputs = model(**inputs)
-        prediction = outputs.logits.argmax(dim=-1).item()
-        intent = int2str(prediction)
-    if intent == "oos":
-        return {"intent": "out of scope (OOS)"}
-    else:
-        intent = intent.replace("_", " ").title()
-        return {"intent": intent}
+def predict_intent_compat(request: Query):
+    """Compatibility endpoint that handles both text and message fields"""
+    try:
+        # Handle both 'text' and 'message' fields for compatibility
+        text = request.message or request.text or ""
+        
+        if not text:
+            return {"error": "No text or message provided"}
+        
+        inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True, max_length=128)
+        with torch.no_grad():
+            outputs = model(**inputs)
+            prediction = outputs.logits.argmax(dim=-1).item()
+            
+            # Debug information
+            print(f"Input: {text}")
+            print(f"Raw prediction index: {prediction}")
+            print(f"Total labels available: {len(INTENT_LABELS)}")
+            
+            intent = int2str(prediction)
+            print(f"Mapped intent: {intent}")
+            
+        if intent == "oos":
+            return {"intent": "out of scope (OOS)"}
+        else:
+            intent = intent.replace("_", " ").title()
+            return {"intent": intent}
+    except Exception as e:
+        print(f"Error in prediction: {e}")
+        return {"intent": "Error", "error": str(e)}
+
+
+
+@app.get("/", response_class=HTMLResponse)
+async def read_root():
+    """Serve the main HTML interface"""
+    html_content = """
+    <!DOCTYPE html>
+    <html lang="en">
+    <head>
+        <meta charset="UTF-8">
+        <title>Intent Classifier Chatbot</title>
+        <meta name="viewport" content="width=device-width, initial-scale=1">
+        <style>
+            body { 
+                font-family: 'Segoe UI', Arial, sans-serif; 
+                margin: 0; 
+                background: #f7f9fa; 
+                color: #222; 
+            }
+            .container { 
+                max-width: 600px; 
+                margin: 60px auto 30px auto; 
+                background: #fff; 
+                border-radius: 12px; 
+                box-shadow: 0 4px 24px rgba(0,0,0,0.08); 
+                padding: 32px 28px 24px 28px;
+            }
+            h1 {
+                text-align: center;
+                color: #2d6cdf;
+                margin-bottom: 18px;
+            }
+            h2 {
+                text-align: center;
+                color: #2d6cdf;
+                margin-bottom: 18px;
+                font-size: 1.5em;
+            }
+            label {
+                font-weight: 500;
+                margin-bottom: 8px;
+                display: block;
+            }
+            textarea {
+                width: 100%;
+                height: 100px;
+                padding: 12px;
+                border: 1px solid #d2d6dc;
+                border-radius: 6px;
+                font-size: 1em;
+                margin-bottom: 18px;
+                box-sizing: border-box;
+                transition: border 0.2s;
+            }
+            textarea:focus {
+                border: 1.5px solid #2d6cdf;
+                outline: none;
+            }
+            button {
+                width: 100%;
+                padding: 12px;
+                background: linear-gradient(90deg, #2d6cdf 60%, #4e9cff 100%);
+                color: #fff;
+                border: none;
+                border-radius: 6px;
+                font-size: 1.1em;
+                font-weight: 600;
+                cursor: pointer;
+                transition: background 0.2s;
+            }
+            button:hover {
+                background: linear-gradient(90deg, #1b4e9b 60%, #3578c7 100%);
+            }
+            .result {
+                margin-top: 24px;
+                font-size: 1.15em;
+                background: #eaf3ff;
+                border-left: 4px solid #2d6cdf;
+                padding: 14px 18px;
+                border-radius: 6px;
+                color: #1a3a5d;
+                word-break: break-word;
+            }
+            .info {
+                margin-top: 18px;
+                font-size: 0.98em;
+                color: #555;
+                background: #f3f6fa;
+                border-radius: 6px;
+                padding: 10px 14px;
+            }
+            footer {
+                margin-top: 40px;
+                text-align: center;
+                color: #888;
+                font-size: 0.97em;
+                padding-bottom: 18px;
+            }
+            @media (max-width: 600px) {
+                .container { padding: 18px 6px 18px 6px; }
+            }
+        </style>
+    </head>
+    <body>
+        <div class="container">
+            <h1>Intent Classifier Chatbot</h1>
+            <h2>Predict User Intent</h2>
+            <div class="info">
+                Enter a message below and click <b>Predict Intent</b> to see what the AI thinks your intent is.<br>
+                <span style="color:#2d6cdf;">Try: <i>"Set an alarm for 7am"</i> or <i>"Transfer money to John"</i></span>
+            </div>
+            <div class="form-group">
+                <label for="message">Your Message:</label>
+                <textarea id="message" placeholder="Type your message here..."></textarea>
+            </div>
+            <button onclick="predictIntent()">Predict Intent</button>
+            <div id="result" class="result" style="display: none;"></div>
+        </div>
+        <footer>
+            Made by <b>Saher Muhamed</b><br>
+            <a href="https://github.com/sahermuhamed1" target="_blank" style="color:#2d6cdf;text-decoration:none;">GitHub</a> &middot; 
+            <a href="mailto:sahermuhamed176@gmail.com" style="color:#2d6cdf;text-decoration:none;">Contact</a>
+        </footer>
+        <script>
+            function predictIntent() {
+                const message = document.getElementById('message').value.trim();
+                const resultDiv = document.getElementById('result');
+                
+                if (!message) {
+                    alert('Please enter a message first!');
+                    return;
+                }
+                
+                resultDiv.style.display = 'block';
+                resultDiv.innerHTML = 'Predicting...';
+                
+                fetch('/predict', {
+                    method: 'POST',
+                    headers: {
+                        'Content-Type': 'application/json',
+                    },
+                    body: JSON.stringify({message: message})
+                })
+                .then(response => response.json())
+                .then(data => {
+                    if (data.error) {
+                        resultDiv.innerHTML = `<span style="color: red;">Error: ${data.error}</span>`;
+                    } else {
+                        resultDiv.innerHTML = `<span style="color: green;">Predicted Intent: ${data.intent || 'Unknown'}</span>`;
+                    }
+                })
+                .catch(error => {
+                    resultDiv.innerHTML = `<span style="color: red;">Error: ${error.message}</span>`;
+                });
+            }
+        </script>
+    </body>
+    </html>
+    """
+    return HTMLResponse(content=html_content)

model/api/start_server.py

@@ -2,3 +2,4 @@ import uvicorn
 
 if __name__ == "__main__":
     uvicorn.run("api:app", host="0.0.0.0", port=8000, reload=True)
+ 

src/fastapi_server.py

@@ -14,4 +14,4 @@ def predict(req: PredictRequest):
         intent = "set_alarm"
     else:
         intent = "unknown"
-    return {"intent": intent}
+    return {"intent": intent}

src/{app.py → main.py}

@@ -1,5 +1,5 @@
 # NOTE: Make sure the FastAPI server is running at http://localhost:8000 before starting this Flask app.
-from flask import Flask, render_template, request
+from flask import Flask, render_template, request, jsonify
 import requests
 
 app = Flask(__name__)
@@ -14,7 +14,7 @@ def index():
         user_text = request.form.get("user_text", "")
         if user_text:
             try:
-                response = requests.post(FASTAPI_URL, json={"text": user_text})
+                response = requests.post(FASTAPI_URL, json={"message": user_text})
                 if response.status_code == 200:
                     prediction = response.json().get("intent", "Unknown")
                 else:
@@ -30,5 +30,28 @@ def index():
                 prediction = f"Error: {str(e)}"
     return render_template("index.html", prediction=prediction, user_text=user_text)
 
+@app.route('/predict', methods=['POST'])
+def predict():
+    try:
+        data = request.get_json()
+        message = data.get('message', '')
+        
+        if not message:
+            return jsonify({'error': 'No message provided'}), 400
+        
+        # Call FastAPI model server
+        response = requests.post(FASTAPI_URL, json={'message': message})
+        
+        if response.status_code == 200:
+            result = response.json()
+            return jsonify({'intent': result.get('intent', 'Unknown')})
+        else:
+            return jsonify({'error': 'Model server error'}), 500
+            
+    except requests.exceptions.ConnectionError:
+        return jsonify({'error': 'Could not connect to FastAPI server. Make sure it\'s running on port 8000.'}), 500
+    except Exception as e:
+        return jsonify({'error': str(e)}), 500
+
 if __name__ == "__main__":
-    app.run(debug=True)
+    app.run(host="0.0.0.0", port=5000, debug=False)

start.sh

@@ -0,0 +1,12 @@
+#!/bin/bash
+
+# Download model files if they don't exist
+if [ ! -d "/app/intent_classifier_model" ]; then
+    echo "Model files not found. Please upload your trained model to the Space."
+    echo "Create the following directories and upload your model files:"
+    echo "- intent_classifier_model/ (containing the trained BERT model)"
+    echo "- intent_classifier_tokenizer/ (containing the tokenizer)"
+fi
+
+# Start the FastAPI application
+exec uvicorn model.api.api:app --host 0.0.0.0 --port 7860

training/workspace.ipynb

@@ -5,16 +5,7 @@
    "execution_count": 1,
    "id": "27ee9040",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/home/saher/miniconda3/envs/AI/lib/python3.9/site-packages/requests/__init__.py:86: RequestsDependencyWarning: Unable to find acceptable character detection dependency (chardet or charset_normalizer).\n",
-      "  warnings.warn(\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "from datasets import load_dataset\n",
     "\n",
@@ -29,6 +20,190 @@
     "test_labels = dataset[\"test\"][\"intent\"]\n"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "78405e22",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['restaurant_reviews',\n",
+       " 'nutrition_info',\n",
+       " 'account_blocked',\n",
+       " 'oil_change_how',\n",
+       " 'time',\n",
+       " 'weather',\n",
+       " 'redeem_rewards',\n",
+       " 'interest_rate',\n",
+       " 'gas_type',\n",
+       " 'accept_reservations',\n",
+       " 'smart_home',\n",
+       " 'user_name',\n",
+       " 'report_lost_card',\n",
+       " 'repeat',\n",
+       " 'whisper_mode',\n",
+       " 'what_are_your_hobbies',\n",
+       " 'order',\n",
+       " 'jump_start',\n",
+       " 'schedule_meeting',\n",
+       " 'meeting_schedule',\n",
+       " 'freeze_account',\n",
+       " 'what_song',\n",
+       " 'meaning_of_life',\n",
+       " 'restaurant_reservation',\n",
+       " 'traffic',\n",
+       " 'make_call',\n",
+       " 'text',\n",
+       " 'bill_balance',\n",
+       " 'improve_credit_score',\n",
+       " 'change_language',\n",
+       " 'no',\n",
+       " 'measurement_conversion',\n",
+       " 'timer',\n",
+       " 'flip_coin',\n",
+       " 'do_you_have_pets',\n",
+       " 'balance',\n",
+       " 'tell_joke',\n",
+       " 'last_maintenance',\n",
+       " 'exchange_rate',\n",
+       " 'uber',\n",
+       " 'car_rental',\n",
+       " 'credit_limit',\n",
+       " 'oos',\n",
+       " 'shopping_list',\n",
+       " 'expiration_date',\n",
+       " 'routing',\n",
+       " 'meal_suggestion',\n",
+       " 'tire_change',\n",
+       " 'todo_list',\n",
+       " 'card_declined',\n",
+       " 'rewards_balance',\n",
+       " 'change_accent',\n",
+       " 'vaccines',\n",
+       " 'reminder_update',\n",
+       " 'food_last',\n",
+       " 'change_ai_name',\n",
+       " 'bill_due',\n",
+       " 'who_do_you_work_for',\n",
+       " 'share_location',\n",
+       " 'international_visa',\n",
+       " 'calendar',\n",
+       " 'translate',\n",
+       " 'carry_on',\n",
+       " 'book_flight',\n",
+       " 'insurance_change',\n",
+       " 'todo_list_update',\n",
+       " 'timezone',\n",
+       " 'cancel_reservation',\n",
+       " 'transactions',\n",
+       " 'credit_score',\n",
+       " 'report_fraud',\n",
+       " 'spending_history',\n",
+       " 'directions',\n",
+       " 'spelling',\n",
+       " 'insurance',\n",
+       " 'what_is_your_name',\n",
+       " 'reminder',\n",
+       " 'where_are_you_from',\n",
+       " 'distance',\n",
+       " 'payday',\n",
+       " 'flight_status',\n",
+       " 'find_phone',\n",
+       " 'greeting',\n",
+       " 'alarm',\n",
+       " 'order_status',\n",
+       " 'confirm_reservation',\n",
+       " 'cook_time',\n",
+       " 'damaged_card',\n",
+       " 'reset_settings',\n",
+       " 'pin_change',\n",
+       " 'replacement_card_duration',\n",
+       " 'new_card',\n",
+       " 'roll_dice',\n",
+       " 'income',\n",
+       " 'taxes',\n",
+       " 'date',\n",
+       " 'who_made_you',\n",
+       " 'pto_request',\n",
+       " 'tire_pressure',\n",
+       " 'how_old_are_you',\n",
+       " 'rollover_401k',\n",
+       " 'pto_request_status',\n",
+       " 'how_busy',\n",
+       " 'application_status',\n",
+       " 'recipe',\n",
+       " 'calendar_update',\n",
+       " 'play_music',\n",
+       " 'yes',\n",
+       " 'direct_deposit',\n",
+       " 'credit_limit_change',\n",
+       " 'gas',\n",
+       " 'pay_bill',\n",
+       " 'ingredients_list',\n",
+       " 'lost_luggage',\n",
+       " 'goodbye',\n",
+       " 'what_can_i_ask_you',\n",
+       " 'book_hotel',\n",
+       " 'are_you_a_bot',\n",
+       " 'next_song',\n",
+       " 'change_speed',\n",
+       " 'plug_type',\n",
+       " 'maybe',\n",
+       " 'w2',\n",
+       " 'oil_change_when',\n",
+       " 'thank_you',\n",
+       " 'shopping_list_update',\n",
+       " 'pto_balance',\n",
+       " 'order_checks',\n",
+       " 'travel_alert',\n",
+       " 'fun_fact',\n",
+       " 'sync_device',\n",
+       " 'schedule_maintenance',\n",
+       " 'apr',\n",
+       " 'transfer',\n",
+       " 'ingredient_substitution',\n",
+       " 'calories',\n",
+       " 'current_location',\n",
+       " 'international_fees',\n",
+       " 'calculator',\n",
+       " 'definition',\n",
+       " 'next_holiday',\n",
+       " 'update_playlist',\n",
+       " 'mpg',\n",
+       " 'min_payment',\n",
+       " 'change_user_name',\n",
+       " 'restaurant_suggestion',\n",
+       " 'travel_notification',\n",
+       " 'cancel',\n",
+       " 'pto_used',\n",
+       " 'travel_suggestion',\n",
+       " 'change_volume']"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "ename": "",
+     "evalue": "",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[1;31mThe Kernel crashed while executing code in the current cell or a previous cell. \n",
+      "\u001b[1;31mPlease review the code in the cell(s) to identify a possible cause of the failure. \n",
+      "\u001b[1;31mClick <a href='https://aka.ms/vscodeJupyterKernelCrash'>here</a> for more info. \n",
+      "\u001b[1;31mView Jupyter <a href='command:jupyter.viewOutput'>log</a> for further details."
+     ]
+    }
+   ],
+   "source": [
+    "# list all the intents with its label string\n",
+    "intents = dataset[\"train\"].features[\"intent\"].names \n",
+    "intents"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 2,