pardhas133 commited on
Commit
fc00054
·
verified ·
1 Parent(s): 5eb92e7

Upload 11 files

Browse files
APP_INTERFACE_GUIDE.md ADDED
@@ -0,0 +1,182 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Streamlit App Interface Guide
2
+
3
+ ## 🚀 Accessing the App
4
+
5
+ The Streamlit app should now be running. Open your web browser and navigate to:
6
+
7
+ **http://localhost:8501**
8
+
9
+ If port 8501 is busy, Streamlit will automatically use the next available port (8502, 8503, etc.). Check the terminal output for the exact URL.
10
+
11
+ ---
12
+
13
+ ## 📱 Interface Overview
14
+
15
+ ### **Main Page Layout**
16
+
17
+ ```
18
+ ┌─────────────────────────────────────────────────────────┐
19
+ │ 🛠️ Engine Predictive Maintenance – Failure Prediction │
20
+ ├─────────────────────────────────────────────────────────┤
21
+ │ │
22
+ │ This app predicts whether an engine is operating │
23
+ │ normally (0) or requires maintenance / at risk of │
24
+ │ failure (1) based on real-time sensor readings. │
25
+ │ │
26
+ │ Adjust the sensor values below and click Predict to │
27
+ │ see the model's classification and the probability of │
28
+ │ a potential fault. │
29
+ │ │
30
+ ├─────────────────────────────────────────────────────────┤
31
+ │ Input Sensor Readings │
32
+ ├─────────────────────────────────────────────────────────┤
33
+ │ │
34
+ │ ┌─────────────────────┐ ┌─────────────────────┐ │
35
+ │ │ Engine RPM │ │ Coolant Pressure │ │
36
+ │ │ [800.0] │ │ [2.0] │ │
37
+ │ │ │ │ │ │
38
+ │ │ Lub Oil Pressure │ │ Lub Oil Temperature │ │
39
+ │ │ [3.0] │ │ [80.0] │ │
40
+ │ │ │ │ │ │
41
+ │ │ Fuel Pressure │ │ Coolant Temperature │ │
42
+ │ │ [10.0] │ │ [80.0] │ │
43
+ │ └─────────────────────┘ └─────────────────────┘ │
44
+ │ │
45
+ │ [Predict Button] │
46
+ │ │
47
+ └─────────────────────────────────────────────────────────┘
48
+
49
+ ┌─────────────────────────────────────────────────────────┐
50
+ │ Sidebar: Model Source │
51
+ ├─────────────────────────────────────────────────────────┤
52
+ │ Load model from: │
53
+ │ ○ Hugging Face Hub │
54
+ │ ● Local file │
55
+ │ │
56
+ │ Note: On Hugging Face Spaces, the model is typically │
57
+ │ loaded from the model hub. Locally, you can choose │
58
+ │ either source as long as you have run the training │
59
+ │ pipeline or configured your HF token. │
60
+ └─────────────────────────────────────────────────────────┘
61
+ ```
62
+
63
+ ---
64
+
65
+ ## 🎯 How to Use the Interface
66
+
67
+ ### **Step 1: Select Model Source (Sidebar)**
68
+ - Choose **"Local file"** if you've run `python src/train.py` locally
69
+ - Choose **"Hugging Face Hub"** if you've uploaded the model to HF and set `HF_TOKEN`
70
+
71
+ ### **Step 2: Enter Sensor Values**
72
+
73
+ Adjust the 6 sensor inputs:
74
+
75
+ 1. **Engine RPM** (0-4000)
76
+ - Default: 800.0
77
+ - Typical range: 500-2000 RPM
78
+
79
+ 2. **Lub Oil Pressure** (0-10 bar/kPa)
80
+ - Default: 3.0
81
+ - Typical range: 2-5 bar
82
+
83
+ 3. **Fuel Pressure** (0-30 bar/kPa)
84
+ - Default: 10.0
85
+ - Typical range: 5-20 bar
86
+
87
+ 4. **Coolant Pressure** (0-10 bar/kPa)
88
+ - Default: 2.0
89
+ - Typical range: 1-4 bar
90
+
91
+ 5. **Lub Oil Temperature** (0-150°C)
92
+ - Default: 80.0
93
+ - Typical range: 70-90°C
94
+
95
+ 6. **Coolant Temperature** (0-150°C)
96
+ - Default: 80.0
97
+ - Typical range: 70-90°C
98
+
99
+ ### **Step 3: Click "Predict"**
100
+
101
+ After clicking the **Predict** button, you'll see one of two results:
102
+
103
+ #### ✅ **Normal Operation**
104
+ ```
105
+ ✅ The engine is LIKELY OPERATING NORMALLY (probability of fault X.XX%).
106
+ ```
107
+
108
+ #### 🚨 **Faulty / Requires Maintenance**
109
+ ```
110
+ 🚨 The engine is LIKELY FAULTY / REQUIRES MAINTENANCE (probability XX.XX%).
111
+ ```
112
+
113
+ ---
114
+
115
+ ## 📊 Example Predictions
116
+
117
+ ### **Example 1: Normal Engine**
118
+ - Engine RPM: 1200
119
+ - Lub Oil Pressure: 3.5
120
+ - Fuel Pressure: 12.0
121
+ - Coolant Pressure: 2.5
122
+ - Lub Oil Temperature: 82.0
123
+ - Coolant Temperature: 85.0
124
+ - **Result**: ✅ Normal operation (low fault probability)
125
+
126
+ ### **Example 2: Faulty Engine**
127
+ - Engine RPM: 400
128
+ - Lub Oil Pressure: 1.5
129
+ - Fuel Pressure: 5.0
130
+ - Coolant Pressure: 1.0
131
+ - Lub Oil Temperature: 95.0
132
+ - Coolant Temperature: 100.0
133
+ - **Result**: 🚨 Requires maintenance (high fault probability)
134
+
135
+ ---
136
+
137
+ ## 🔧 Troubleshooting
138
+
139
+ ### **App Not Loading?**
140
+ 1. Check terminal for errors
141
+ 2. Verify port 8501 is available: `lsof -ti:8501`
142
+ 3. Try a different port: `streamlit run src/app.py --server.port 8502`
143
+
144
+ ### **Model Not Found Error?**
145
+ 1. **For Local**: Run `python src/train.py` first to create `models/best_model.joblib`
146
+ 2. **For HF**: Set `HF_TOKEN` and `HF_MODEL_REPO` environment variables
147
+
148
+ ### **Import Errors?**
149
+ 1. Activate virtual environment: `source .venv/bin/activate`
150
+ 2. Install dependencies: `pip install -r requirements.txt`
151
+
152
+ ---
153
+
154
+ ## 🎨 Interface Features
155
+
156
+ - **Clean, centered layout** for easy input
157
+ - **Two-column form** for organized sensor inputs
158
+ - **Real-time prediction** with probability scores
159
+ - **Color-coded results**: Green for normal, Red for faulty
160
+ - **Sidebar model selection** for flexibility
161
+ - **Responsive design** that works on different screen sizes
162
+
163
+ ---
164
+
165
+ ## 📸 Screenshots for Your Report
166
+
167
+ When documenting this in your final report, you can:
168
+ 1. Take a screenshot of the input form
169
+ 2. Take a screenshot showing a "Normal" prediction
170
+ 3. Take a screenshot showing a "Faulty" prediction
171
+ 4. Include the URL: `http://localhost:8501` (or your deployed HF Space URL)
172
+
173
+ ---
174
+
175
+ ## 🚀 Next Steps
176
+
177
+ 1. **Test the app locally** with different sensor values
178
+ 2. **Deploy to Hugging Face Space** using `python src/deploy_to_hf.py`
179
+ 3. **Include screenshots** in your final report/notebook
180
+ 4. **Document the interface** in your submission
181
+
182
+ The app is ready to use! 🎉
CONFIGURATION_GUIDE.md ADDED
@@ -0,0 +1,153 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Configuration Guide: Hugging Face & GitHub Setup
2
+
3
+ This guide shows you exactly where to configure your Hugging Face Space details and GitHub repository information.
4
+
5
+ ---
6
+
7
+ ## 1. Hugging Face Configuration
8
+
9
+ ### Option A: Update `src/config.py` (Recommended for Local Testing)
10
+
11
+ Edit `/Users/ananttripathi/Desktop/mlops/src/config.py` and replace the placeholder values:
12
+
13
+ ```python
14
+ # Lines 58-62 in src/config.py
15
+ HF_DATASET_REPO = os.getenv(
16
+ "HF_DATASET_REPO", "your-username/engine-maintenance-dataset" # <-- Replace "your-username"
17
+ )
18
+ HF_MODEL_REPO = os.getenv("HF_MODEL_REPO", "your-username/engine-maintenance-model") # <-- Replace
19
+ HF_SPACE_REPO = os.getenv("HF_SPACE_REPO", "your-username/engine-maintenance-space") # <-- Replace
20
+ ```
21
+
22
+ **Example** (HF username: `ananttripathiak`, GitHub username: `ananttripathi`):
23
+ ```python
24
+ HF_DATASET_REPO = os.getenv(
25
+ "HF_DATASET_REPO", "ananttripathiak/engine-maintenance-dataset"
26
+ )
27
+ HF_MODEL_REPO = os.getenv("HF_MODEL_REPO", "ananttripathiak/engine-maintenance-model")
28
+ HF_SPACE_REPO = os.getenv("HF_SPACE_REPO", "ananttripathiak/engine-maintenance-space")
29
+ ```
30
+
31
+ ### Option B: Set Environment Variables (For Local Testing)
32
+
33
+ Instead of editing `config.py`, you can export these in your terminal:
34
+
35
+ ```bash
36
+ export HF_TOKEN="hf_your_token_here"
37
+ export HF_DATASET_REPO="your-username/engine-maintenance-dataset"
38
+ export HF_MODEL_REPO="your-username/engine-maintenance-model"
39
+ export HF_SPACE_REPO="your-username/engine-maintenance-space"
40
+ ```
41
+
42
+ ---
43
+
44
+ ## 2. GitHub Repository Configuration
45
+
46
+ ### A. GitHub Secrets (Required for GitHub Actions)
47
+
48
+ Go to your GitHub repository → **Settings** → **Secrets and variables** → **Actions** → **New repository secret**
49
+
50
+ Add these 4 secrets:
51
+
52
+ 1. **`HF_TOKEN`**
53
+ - Value: Your Hugging Face access token (get it from https://huggingface.co/settings/tokens)
54
+ - Example: `hf_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx`
55
+
56
+ 2. **`HF_DATASET_REPO`**
57
+ - Value: Your Hugging Face dataset repo ID
58
+ - Example: `ananttripathiak/engine-maintenance-dataset`
59
+
60
+ 3. **`HF_MODEL_REPO`**
61
+ - Value: Your Hugging Face model repo ID
62
+ - Example: `ananttripathiak/engine-maintenance-model`
63
+
64
+ 4. **`HF_SPACE_REPO`**
65
+ - Value: Your Hugging Face Space repo ID
66
+ - Example: `ananttripathiak/engine-maintenance-space`
67
+
68
+ **Note:** The GitHub Actions workflow (`.github/workflows/pipeline.yml`) automatically reads these secrets. No code changes needed!
69
+
70
+ ### B. Update README.md with Your GitHub Repo URL
71
+
72
+ Edit `/Users/ananttripathi/Desktop/mlops/README.md` and add your GitHub repository link:
73
+
74
+ ```markdown
75
+ ## GitHub Repository
76
+
77
+ - **Repository URL**: https://github.com/your-username/engine-predictive-maintenance
78
+ - **GitHub Actions**: https://github.com/your-username/engine-predictive-maintenance/actions
79
+ ```
80
+
81
+ ---
82
+
83
+ ## 3. Quick Setup Checklist
84
+
85
+ - [ ] **Hugging Face Account**: Create account at https://huggingface.co
86
+ - [ ] **HF Access Token**: Generate at https://huggingface.co/settings/tokens (needs `write` permission)
87
+ - [ ] **Update `src/config.py`**: Replace `"username"` with your actual HF username
88
+ - [ ] **Create HF Repos** (optional - scripts will create them automatically):
89
+ - Dataset repo: `your-username/engine-maintenance-dataset`
90
+ - Model repo: `your-username/engine-maintenance-model`
91
+ - Space repo: `your-username/engine-maintenance-space`
92
+ - [ ] **GitHub Repository**: Create a new repo and push this `mlops` folder
93
+ - [ ] **GitHub Secrets**: Add the 4 secrets listed above in your GitHub repo settings
94
+ - [ ] **Test Locally**: Run `python src/data_register.py` to verify HF connection
95
+
96
+ ---
97
+
98
+ ## 4. Testing Your Configuration
99
+
100
+ ### Test Local Configuration
101
+
102
+ ```bash
103
+ cd /Users/ananttripathi/Desktop/mlops
104
+ source .venv/bin/activate
105
+
106
+ # Set your HF token (if not in config.py)
107
+ export HF_TOKEN="hf_your_token_here"
108
+
109
+ # Test data registration
110
+ python src/data_register.py
111
+
112
+ # Test data preparation
113
+ python src/data_prep.py
114
+
115
+ # Test model training
116
+ python src/train.py
117
+
118
+ # Test deployment
119
+ python src/deploy_to_hf.py
120
+ ```
121
+
122
+ ### Test GitHub Actions
123
+
124
+ 1. Push your code to GitHub:
125
+ ```bash
126
+ git add .
127
+ git commit -m "Initial commit: Predictive maintenance MLOps pipeline"
128
+ git push origin main
129
+ ```
130
+
131
+ 2. Go to your GitHub repo → **Actions** tab
132
+ 3. You should see the "Predictive Maintenance Pipeline" workflow running
133
+ 4. All 4 jobs should complete successfully (green checkmarks)
134
+
135
+ ---
136
+
137
+ ## 5. Where Each Configuration is Used
138
+
139
+ | Configuration | Used In | Purpose |
140
+ |--------------|---------|---------|
141
+ | `HF_DATASET_REPO` | `src/data_register.py`, `src/data_prep.py`, `src/train.py` | Dataset storage and retrieval |
142
+ | `HF_MODEL_REPO` | `src/train.py`, `src/inference.py`, `src/app.py` | Model storage and loading |
143
+ | `HF_SPACE_REPO` | `src/deploy_to_hf.py` | Streamlit app deployment |
144
+ | `HF_TOKEN` | All HF-related scripts | Authentication |
145
+ | GitHub Secrets | `.github/workflows/pipeline.yml` | CI/CD automation |
146
+
147
+ ---
148
+
149
+ ## Need Help?
150
+
151
+ - **Hugging Face Docs**: https://huggingface.co/docs/hub
152
+ - **GitHub Actions Docs**: https://docs.github.com/en/actions
153
+ - **GitHub Secrets**: https://docs.github.com/en/actions/security-guides/encrypted-secrets
Dockerfile ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ FROM python:3.10-slim
2
+
3
+ WORKDIR /app
4
+
5
+ ENV PYTHONUNBUFFERED=1 \
6
+ PIP_NO_CACHE_DIR=1
7
+
8
+ # Install system dependencies (if needed for some Python packages)
9
+ RUN apt-get update && apt-get install -y --no-install-recommends \
10
+ build-essential \
11
+ && rm -rf /var/lib/apt/lists/*
12
+
13
+ COPY requirements.txt .
14
+ RUN pip install --upgrade pip && pip install -r requirements.txt
15
+
16
+ COPY . .
17
+
18
+ EXPOSE 7860
19
+
20
+ CMD ["streamlit", "run", "src/app.py", "--server.port", "7860", "--server.address", "0.0.0.0"]
21
+
GITHUB_REPO_INFO.md ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # GitHub Repository Information
2
+
3
+ ## Repository Details
4
+
5
+ - **GitHub Username**: `ananttripathi`
6
+ - **Repository Name**: `engine-predictive-maintenance` (or your chosen name)
7
+ - **Full Repository URL**: `https://github.com/ananttripathi/engine-predictive-maintenance`
8
+
9
+ ## Where to Add This Information
10
+
11
+ ### 1. In Your Final Notebook/Report
12
+
13
+ Add a section like this:
14
+
15
+ ```markdown
16
+ ## GitHub Repository
17
+
18
+ - **Repository URL**: https://github.com/ananttripathi/engine-predictive-maintenance
19
+ - **GitHub Actions Workflow**: https://github.com/ananttripathi/engine-predictive-maintenance/actions
20
+ - **Repository Structure**: [Screenshot of folder structure]
21
+ - **Workflow Execution**: [Screenshot of successful pipeline runs]
22
+ ```
23
+
24
+ ### 2. In README.md
25
+
26
+ The README already includes a placeholder. After you create the repo, update it with your actual repo name.
27
+
28
+ ### 3. GitHub Actions Workflow
29
+
30
+ **No changes needed!** The `.github/workflows/pipeline.yml` file automatically runs in the context of whatever repository it's pushed to. It doesn't need the repo URL hardcoded.
31
+
32
+ ## Steps to Set Up GitHub Repository
33
+
34
+ 1. **Create the repository on GitHub:**
35
+ - Go to https://github.com/new
36
+ - Repository name: `engine-predictive-maintenance` (or your choice)
37
+ - Description: "Predictive Maintenance MLOps Pipeline for Engine Failure Classification"
38
+ - Choose Public or Private
39
+ - **Don't** initialize with README, .gitignore, or license (we already have files)
40
+
41
+ 2. **Push your code:**
42
+ ```bash
43
+ cd /Users/ananttripathi/Desktop/mlops
44
+ git init
45
+ git add .
46
+ git commit -m "Initial commit: Predictive maintenance MLOps pipeline"
47
+ git branch -M main
48
+ git remote add origin https://github.com/ananttripathi/engine-predictive-maintenance.git
49
+ git push -u origin main
50
+ ```
51
+
52
+ 3. **Add GitHub Secrets:**
53
+ - Go to: https://github.com/ananttripathi/engine-predictive-maintenance/settings/secrets/actions
54
+ - Click "New repository secret"
55
+ - Add these 4 secrets:
56
+ - `HF_TOKEN` → Your Hugging Face token
57
+ - `HF_DATASET_REPO` → `ananttripathiak/engine-maintenance-dataset`
58
+ - `HF_MODEL_REPO` → `ananttripathiak/engine-maintenance-model`
59
+ - `HF_SPACE_REPO` → `ananttripathiak/engine-maintenance-space`
60
+
61
+ 4. **Verify GitHub Actions:**
62
+ - After pushing, go to: https://github.com/ananttripathi/engine-predictive-maintenance/actions
63
+ - You should see the "Predictive Maintenance Pipeline" workflow running
64
+ - All 4 jobs should complete successfully
65
+
66
+ ## Important Notes
67
+
68
+ - **GitHub username** (`ananttripathi`) is used for the repository URL
69
+ - **Hugging Face username** might be different - check your HF account and update `src/config.py` if needed
70
+ - The GitHub Actions workflow reads secrets automatically - no code changes needed
71
+ - Repository name can be anything you want (e.g., `engine-predictive-maintenance`, `mlops-project`, etc.)
MANUAL_VS_AUTOMATIC.md ADDED
@@ -0,0 +1,235 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Manual vs Automatic Uploads - Quick Guide
2
+
3
+ ## 🚀 AUTOMATIC Uploads (Via Scripts)
4
+
5
+ These happen automatically when you run the scripts. **You don't need to manually upload anything to Hugging Face.**
6
+
7
+ ### ✅ Automatic: Hugging Face Dataset Repo
8
+ **Script**: `python src/data_register.py` and `python src/data_prep.py`
9
+
10
+ **What gets uploaded automatically:**
11
+ - ✅ `data/engine_data.csv` → HF Dataset Repo
12
+ - ✅ `data/processed/train.csv` → HF Dataset Repo
13
+ - ✅ `data/processed/test.csv` → HF Dataset Repo
14
+
15
+ **You just run:**
16
+ ```bash
17
+ python src/data_register.py # Auto-uploads raw data
18
+ python src/data_prep.py # Auto-uploads train/test
19
+ ```
20
+
21
+ **That's it!** The scripts handle everything - creating the repo, uploading files, etc.
22
+
23
+ ---
24
+
25
+ ### ✅ Automatic: Hugging Face Model Repo
26
+ **Script**: `python src/train.py`
27
+
28
+ **What gets uploaded automatically:**
29
+ - ✅ `models/best_model.joblib` → HF Model Repo (as `model.joblib`)
30
+
31
+ **You just run:**
32
+ ```bash
33
+ python src/train.py
34
+ ```
35
+
36
+ **That's it!** The script:
37
+ 1. Trains the model
38
+ 2. Saves it locally
39
+ 3. Automatically uploads to HF Model Repo
40
+
41
+ ---
42
+
43
+ ### ✅ Automatic: Hugging Face Space (Streamlit App)
44
+ **Script**: `python src/deploy_to_hf.py`
45
+
46
+ **What gets uploaded automatically:**
47
+ - ✅ `src/app.py` (Streamlit app)
48
+ - ✅ `src/inference.py`
49
+ - ✅ `src/config.py`
50
+ - ✅ `Dockerfile`
51
+ - ✅ `requirements.txt`
52
+ - ✅ Other `src/*.py` files needed
53
+
54
+ **You just run:**
55
+ ```bash
56
+ python src/deploy_to_hf.py
57
+ ```
58
+
59
+ **That's it!** The script:
60
+ 1. Creates/updates the HF Space
61
+ 2. Uploads all deployment files
62
+ 3. Configures it as a Streamlit app
63
+
64
+ ---
65
+
66
+ ## 📤 MANUAL Uploads (You Do This)
67
+
68
+ Only **ONE** thing needs to be done manually:
69
+
70
+ ### ⚠️ Manual: GitHub Repository
71
+
72
+ **What you need to do manually:**
73
+ - Push your entire `mlops/` folder to GitHub
74
+
75
+ **Steps:**
76
+ ```bash
77
+ cd /Users/ananttripathi/Desktop/mlops
78
+
79
+ # 1. Initialize git (if not done)
80
+ git init
81
+
82
+ # 2. Create .gitignore (to exclude large files)
83
+ cat > .gitignore << EOF
84
+ .venv/
85
+ __pycache__/
86
+ *.pyc
87
+ mlruns/
88
+ *.log
89
+ .DS_Store
90
+ EOF
91
+
92
+ # 3. Add all files
93
+ git add .
94
+
95
+ # 4. Commit
96
+ git commit -m "Initial commit: Predictive Maintenance MLOps Pipeline"
97
+
98
+ # 5. Add your GitHub repo as remote
99
+ git remote add origin https://github.com/ananttripathi/engine-predictive-maintenance.git
100
+
101
+ # 6. Push to GitHub
102
+ git push -u origin main
103
+ ```
104
+
105
+ **That's the ONLY manual upload!**
106
+
107
+ ---
108
+
109
+ ## 📊 Summary Table
110
+
111
+ | Destination | Upload Method | What You Do |
112
+ |------------|---------------|-------------|
113
+ | **GitHub** | ⚠️ **MANUAL** | Run `git push` commands |
114
+ | **HF Dataset** | ✅ **AUTOMATIC** | Run `python src/data_register.py` and `python src/data_prep.py` |
115
+ | **HF Model** | ✅ **AUTOMATIC** | Run `python src/train.py` |
116
+ | **HF Space** | ✅ **AUTOMATIC** | Run `python src/deploy_to_hf.py` |
117
+
118
+ ---
119
+
120
+ ## 🎯 Complete Workflow
121
+
122
+ ### Step 1: Manual - GitHub (Do This First)
123
+ ```bash
124
+ # Push everything to GitHub
125
+ git init
126
+ git add .
127
+ git commit -m "Initial commit"
128
+ git remote add origin https://github.com/ananttripathi/engine-predictive-maintenance.git
129
+ git push -u origin main
130
+ ```
131
+ ⏱️ **Time**: 2-3 minutes
132
+
133
+ ---
134
+
135
+ ### Step 2: Automatic - HF Dataset
136
+ ```bash
137
+ export HF_TOKEN="your_hf_token"
138
+ python src/data_register.py
139
+ python src/data_prep.py
140
+ ```
141
+ ⏱️ **Time**: 1-2 minutes (scripts do everything)
142
+
143
+ ---
144
+
145
+ ### Step 3: Automatic - HF Model
146
+ ```bash
147
+ python src/train.py
148
+ ```
149
+ ⏱️ **Time**: 5-10 minutes (trains model + auto-uploads)
150
+
151
+ ---
152
+
153
+ ### Step 4: Automatic - HF Space
154
+ ```bash
155
+ python src/deploy_to_hf.py
156
+ ```
157
+ ⏱️ **Time**: 1-2 minutes (scripts do everything)
158
+
159
+ ---
160
+
161
+ ## ✅ What You Need to Prepare
162
+
163
+ ### Before Running Scripts:
164
+
165
+ 1. **Set Environment Variable:**
166
+ ```bash
167
+ export HF_TOKEN="hf_your_token_here"
168
+ ```
169
+
170
+ 2. **Update `src/config.py`** (already done):
171
+ - HF repos are set to `ananttripathiak/...`
172
+
173
+ 3. **That's it!** Scripts handle the rest automatically.
174
+
175
+ ---
176
+
177
+ ## 🔍 Verification
178
+
179
+ After running scripts, verify automatic uploads:
180
+
181
+ 1. **HF Dataset**: https://huggingface.co/datasets/ananttripathiak/engine-maintenance-dataset
182
+ - Should see: `data/engine_data.csv`, `data/train.csv`, `data/test.csv`
183
+
184
+ 2. **HF Model**: https://huggingface.co/ananttripathiak/engine-maintenance-model
185
+ - Should see: `model.joblib`
186
+
187
+ 3. **HF Space**: https://huggingface.co/spaces/ananttripathiak/engine-maintenance-space
188
+ - Should see: `src/app.py`, `Dockerfile`, `requirements.txt`
189
+ - App should be running!
190
+
191
+ ---
192
+
193
+ ## 💡 Key Points
194
+
195
+ ✅ **Automatic (3 out of 4):**
196
+ - HF Dataset upload
197
+ - HF Model upload
198
+ - HF Space deployment
199
+
200
+ ⚠️ **Manual (1 out of 4):**
201
+ - GitHub push (only this one!)
202
+
203
+ 🎯 **Bottom Line:**
204
+ - **75% automatic** - Just run the scripts!
205
+ - **25% manual** - Just push to GitHub once!
206
+
207
+ ---
208
+
209
+ ## 🚨 Common Questions
210
+
211
+ **Q: Do I need to manually create HF repos?**
212
+ A: **No!** Scripts create them automatically if they don't exist.
213
+
214
+ **Q: Do I need to manually upload files to HF?**
215
+ A: **No!** Scripts upload everything automatically.
216
+
217
+ **Q: What if a script fails?**
218
+ A: Check `HF_TOKEN` is set, then re-run the script. It will continue from where it left off.
219
+
220
+ **Q: Can I skip GitHub and just use HF?**
221
+ A: For the project rubric, you need GitHub for the CI/CD workflow. But for HF-only, yes, scripts handle everything.
222
+
223
+ ---
224
+
225
+ ## 📝 Quick Checklist
226
+
227
+ - [ ] **Manual**: Push to GitHub (`git push`)
228
+ - [ ] **Automatic**: Run `python src/data_register.py`
229
+ - [ ] **Automatic**: Run `python src/data_prep.py`
230
+ - [ ] **Automatic**: Run `python src/train.py`
231
+ - [ ] **Automatic**: Run `python src/deploy_to_hf.py`
232
+ - [ ] Verify all uploads completed successfully
233
+
234
+ **Total Manual Work: ~5 minutes (just GitHub)**
235
+ **Total Automatic Work: ~10-15 minutes (scripts do everything)**
Notebook_AML_and_MLOps_Project%20%282%29.ipynb ADDED
@@ -0,0 +1,1808 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "markdown",
5
+ "metadata": {
6
+ "id": "klg2JF-oBblG"
7
+ },
8
+ "source": [
9
+ "# Problem Statement"
10
+ ]
11
+ },
12
+ {
13
+ "cell_type": "markdown",
14
+ "metadata": {
15
+ "id": "m0CcOjZ-BblL"
16
+ },
17
+ "source": [
18
+ "## **Business Context**"
19
+ ]
20
+ },
21
+ {
22
+ "cell_type": "markdown",
23
+ "metadata": {
24
+ "id": "uyT6Koe7BblM"
25
+ },
26
+ "source": [
27
+ "\"Visit with Us,\" a leading travel company, is revolutionizing the tourism industry by leveraging data-driven strategies to optimize operations and customer engagement. While introducing a new package offering, such as the Wellness Tourism Package, the company faces challenges in targeting the right customers efficiently. The manual approach to identifying potential customers is inconsistent, time-consuming, and prone to errors, leading to missed opportunities and suboptimal campaign performance.\n",
28
+ "\n",
29
+ "To address these issues, the company aims to implement a scalable and automated system that integrates customer data, predicts potential buyers, and enhances decision-making for marketing strategies. By utilizing an MLOps pipeline, the company seeks to achieve seamless integration of data preprocessing, model development, deployment, and CI/CD practices for continuous improvement. This system will ensure efficient targeting of customers, timely updates to the predictive model, and adaptation to evolving customer behaviors, ultimately driving growth and customer satisfaction.\n"
30
+ ]
31
+ },
32
+ {
33
+ "cell_type": "markdown",
34
+ "metadata": {
35
+ "id": "zm6bNQOJBblO"
36
+ },
37
+ "source": [
38
+ "## **Objective**"
39
+ ]
40
+ },
41
+ {
42
+ "cell_type": "markdown",
43
+ "metadata": {
44
+ "id": "7PYtjk_YBblO"
45
+ },
46
+ "source": [
47
+ "As an MLOps Engineer at \"Visit with Us,\" your responsibility is to design and deploy an MLOps pipeline on GitHub to automate the end-to-end workflow for predicting customer purchases. The primary objective is to build a model that predicts whether a customer will purchase the newly introduced Wellness Tourism Package before contacting them. The pipeline will include data cleaning, preprocessing, transformation, model building, training, evaluation, and deployment, ensuring consistent performance and scalability. By leveraging GitHub Actions for CI/CD integration, the system will enable automated updates, streamline model deployment, and improve operational efficiency. This robust predictive solution will empower policymakers to make data-driven decisions, enhance marketing strategies, and effectively target potential customers, thereby driving customer acquisition and business growth."
48
+ ]
49
+ },
50
+ {
51
+ "cell_type": "markdown",
52
+ "metadata": {
53
+ "id": "z8C11AzTBblP"
54
+ },
55
+ "source": [
56
+ "## **Data Description**"
57
+ ]
58
+ },
59
+ {
60
+ "cell_type": "markdown",
61
+ "metadata": {
62
+ "id": "9DQx3pkaBblP"
63
+ },
64
+ "source": [
65
+ "The dataset contains customer and interaction data that serve as key attributes for predicting the likelihood of purchasing the Wellness Tourism Package. The detailed attributes are:\n",
66
+ "\n",
67
+ "**Customer Details**\n",
68
+ "- **CustomerID:** Unique identifier for each customer.\n",
69
+ "- **ProdTaken:** Target variable indicating whether the customer has purchased a package (0: No, 1: Yes).\n",
70
+ "- **Age:** Age of the customer.\n",
71
+ "- **TypeofContact:** The method by which the customer was contacted (Company Invited or Self Inquiry).\n",
72
+ "- **CityTier:** The city category based on development, population, and living standards (Tier 1 > Tier 2 > Tier 3).\n",
73
+ "- **Occupation:** Customer's occupation (e.g., Salaried, Freelancer).\n",
74
+ "- **Gender:** Gender of the customer (Male, Female).\n",
75
+ "- **NumberOfPersonVisiting:** Total number of people accompanying the customer on the trip.\n",
76
+ "- **PreferredPropertyStar:** Preferred hotel rating by the customer.\n",
77
+ "- **MaritalStatus:** Marital status of the customer (Single, Married, Divorced).\n",
78
+ "- **NumberOfTrips:** Average number of trips the customer takes annually.\n",
79
+ "- **Passport:** Whether the customer holds a valid passport (0: No, 1: Yes).\n",
80
+ "- **OwnCar:** Whether the customer owns a car (0: No, 1: Yes).\n",
81
+ "- **NumberOfChildrenVisiting:** Number of children below age 5 accompanying the customer.\n",
82
+ "- **Designation:** Customer's designation in their current organization.\n",
83
+ "- **MonthlyIncome:** Gross monthly income of the customer.\n",
84
+ "\n",
85
+ "**Customer Interaction Data**\n",
86
+ "- **PitchSatisfactionScore:** Score indicating the customer's satisfaction with the sales pitch.\n",
87
+ "- **ProductPitched:** The type of product pitched to the customer.\n",
88
+ "- **NumberOfFollowups:** Total number of follow-ups by the salesperson after the sales pitch.-\n",
89
+ "- **DurationOfPitch:** Duration of the sales pitch delivered to the customer.\n"
90
+ ]
91
+ },
92
+ {
93
+ "cell_type": "markdown",
94
+ "metadata": {
95
+ "id": "0LbSu_p2jYfe"
96
+ },
97
+ "source": [
98
+ "# Model Building"
99
+ ]
100
+ },
101
+ {
102
+ "cell_type": "code",
103
+ "execution_count": 1,
104
+ "metadata": {
105
+ "colab": {
106
+ "base_uri": "https://localhost:8080/"
107
+ },
108
+ "executionInfo": {
109
+ "elapsed": 15,
110
+ "status": "ok",
111
+ "timestamp": 1765705284123,
112
+ "user": {
113
+ "displayName": "Anant Tripathi",
114
+ "userId": "05588283814303116545"
115
+ },
116
+ "user_tz": -330
117
+ },
118
+ "id": "giodc4KknHID",
119
+ "outputId": "2caa2d21-8ef6-4259-8d79-5573ee1372d1"
120
+ },
121
+ "outputs": [
122
+ {
123
+ "name": "stdout",
124
+ "output_type": "stream",
125
+ "text": [
126
+ "/content\n"
127
+ ]
128
+ }
129
+ ],
130
+ "source": [
131
+ "# Create a master folder to keep all files created when executing the below code cells\n",
132
+ "import os\n",
133
+ "print(os.getcwd())\n",
134
+ "os.makedirs(\"tourism_project\", exist_ok=True)"
135
+ ]
136
+ },
137
+ {
138
+ "cell_type": "code",
139
+ "execution_count": 2,
140
+ "metadata": {
141
+ "executionInfo": {
142
+ "elapsed": 1,
143
+ "status": "ok",
144
+ "timestamp": 1765705284135,
145
+ "user": {
146
+ "displayName": "Anant Tripathi",
147
+ "userId": "05588283814303116545"
148
+ },
149
+ "user_tz": -330
150
+ },
151
+ "id": "SUKPoy0EA4jj"
152
+ },
153
+ "outputs": [],
154
+ "source": [
155
+ "# Create a folder for storing the model building files\n",
156
+ "os.makedirs(\"tourism_project/model_building\", exist_ok=True)"
157
+ ]
158
+ },
159
+ {
160
+ "cell_type": "markdown",
161
+ "metadata": {
162
+ "id": "9DtS3gNDjBbR"
163
+ },
164
+ "source": [
165
+ "## Data Registration"
166
+ ]
167
+ },
168
+ {
169
+ "cell_type": "code",
170
+ "execution_count": 3,
171
+ "metadata": {
172
+ "executionInfo": {
173
+ "elapsed": 10,
174
+ "status": "ok",
175
+ "timestamp": 1765705284146,
176
+ "user": {
177
+ "displayName": "Anant Tripathi",
178
+ "userId": "05588283814303116545"
179
+ },
180
+ "user_tz": -330
181
+ },
182
+ "id": "ZagOeVxJOtJ9"
183
+ },
184
+ "outputs": [],
185
+ "source": [
186
+ "os.makedirs(\"tourism_project/data\", exist_ok=True)"
187
+ ]
188
+ },
189
+ {
190
+ "cell_type": "markdown",
191
+ "metadata": {
192
+ "id": "WxXiD9ZXxodF"
193
+ },
194
+ "source": [
195
+ "Once the **data** folder created after executing the above cell, please upload the **tourism.csv** in to the folder"
196
+ ]
197
+ },
198
+ {
199
+ "cell_type": "markdown",
200
+ "metadata": {
201
+ "id": "hh2TjRG5WJ4Z"
202
+ },
203
+ "source": [
204
+ "## Data Preparation"
205
+ ]
206
+ },
207
+ {
208
+ "cell_type": "code",
209
+ "execution_count": 4,
210
+ "metadata": {
211
+ "colab": {
212
+ "base_uri": "https://localhost:8080/"
213
+ },
214
+ "executionInfo": {
215
+ "elapsed": 9,
216
+ "status": "ok",
217
+ "timestamp": 1765705284156,
218
+ "user": {
219
+ "displayName": "Anant Tripathi",
220
+ "userId": "05588283814303116545"
221
+ },
222
+ "user_tz": -330
223
+ },
224
+ "id": "EHVRGAeoOtJ-",
225
+ "outputId": "016d1a7d-e728-4d1e-a208-c4a08c9213a4"
226
+ },
227
+ "outputs": [
228
+ {
229
+ "name": "stdout",
230
+ "output_type": "stream",
231
+ "text": [
232
+ "Writing tourism_project/model_building/data_register.py\n"
233
+ ]
234
+ }
235
+ ],
236
+ "source": [
237
+ "%%writefile tourism_project/model_building/data_register.py\n",
238
+ "from huggingface_hub.utils import RepositoryNotFoundError, HfHubHTTPError\n",
239
+ "from huggingface_hub import HfApi, create_repo\n",
240
+ "import os\n",
241
+ "\n",
242
+ "\n",
243
+ "repo_id = \"ananttripathiak/tourism-dataset\"\n",
244
+ "repo_type = \"dataset\"\n",
245
+ "\n",
246
+ "# Initialize API client\n",
247
+ "api = HfApi(token=os.getenv(\"HF_TOKEN\"))\n",
248
+ "\n",
249
+ "# Step 1: Check if the space exists\n",
250
+ "try:\n",
251
+ " api.repo_info(repo_id=repo_id, repo_type=repo_type)\n",
252
+ " print(f\"Space '{repo_id}' already exists. Using it.\")\n",
253
+ "except RepositoryNotFoundError:\n",
254
+ " print(f\"Space '{repo_id}' not found. Creating new space...\")\n",
255
+ " create_repo(repo_id=repo_id, repo_type=repo_type, private=False)\n",
256
+ " print(f\"Space '{repo_id}' created.\")\n",
257
+ "\n",
258
+ "api.upload_folder(\n",
259
+ " folder_path=\"tourism_project/data\",\n",
260
+ " repo_id=repo_id,\n",
261
+ " repo_type=repo_type,\n",
262
+ ")"
263
+ ]
264
+ },
265
+ {
266
+ "cell_type": "markdown",
267
+ "metadata": {
268
+ "id": "eZZKnLkLjeM4"
269
+ },
270
+ "source": [
271
+ "## Model Training and Registration with Experimentation Tracking"
272
+ ]
273
+ },
274
+ {
275
+ "cell_type": "code",
276
+ "execution_count": 5,
277
+ "metadata": {
278
+ "executionInfo": {
279
+ "elapsed": 2,
280
+ "status": "ok",
281
+ "timestamp": 1765705284159,
282
+ "user": {
283
+ "displayName": "Anant Tripathi",
284
+ "userId": "05588283814303116545"
285
+ },
286
+ "user_tz": -330
287
+ },
288
+ "id": "LFmrcXT_OtJ-"
289
+ },
290
+ "outputs": [],
291
+ "source": [
292
+ "# !pip install mlflow scikit-learn huggingface_hub\n",
293
+ "# import sys\n",
294
+ "# !{sys.executable} -m pip install mlflow"
295
+ ]
296
+ },
297
+ {
298
+ "cell_type": "code",
299
+ "execution_count": 6,
300
+ "metadata": {
301
+ "colab": {
302
+ "base_uri": "https://localhost:8080/"
303
+ },
304
+ "executionInfo": {
305
+ "elapsed": 26,
306
+ "status": "ok",
307
+ "timestamp": 1765705284186,
308
+ "user": {
309
+ "displayName": "Anant Tripathi",
310
+ "userId": "05588283814303116545"
311
+ },
312
+ "user_tz": -330
313
+ },
314
+ "id": "DatpH_YdOtJ-",
315
+ "outputId": "3f7f7b88-26e9-4e36-df88-b83e5f28fd09"
316
+ },
317
+ "outputs": [
318
+ {
319
+ "name": "stdout",
320
+ "output_type": "stream",
321
+ "text": [
322
+ "Writing tourism_project/model_building/train.py\n"
323
+ ]
324
+ }
325
+ ],
326
+ "source": [
327
+ "%%writefile tourism_project/model_building/train.py\n",
328
+ "\n",
329
+ "# for data manipulation\n",
330
+ "import pandas as pd\n",
331
+ "import numpy as np\n",
332
+ "# for data preprocessing and pipeline creation\n",
333
+ "from sklearn.preprocessing import StandardScaler\n",
334
+ "from sklearn.compose import make_column_transformer\n",
335
+ "from sklearn.pipeline import make_pipeline\n",
336
+ "# for model training, tuning, and evaluation\n",
337
+ "import xgboost as xgb\n",
338
+ "from sklearn.model_selection import GridSearchCV\n",
339
+ "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, roc_auc_score, classification_report, confusion_matrix\n",
340
+ "# for model serialization\n",
341
+ "import joblib\n",
342
+ "# for creating a folder\n",
343
+ "import os\n",
344
+ "# for hugging face space authentication to upload files\n",
345
+ "from huggingface_hub import login, HfApi, create_repo\n",
346
+ "from huggingface_hub.utils import RepositoryNotFoundError, HfHubHTTPError\n",
347
+ "import mlflow\n",
348
+ "\n",
349
+ "# Set up MLflow tracking\n",
350
+ "mlflow.set_tracking_uri(\"http://localhost:5000\")\n",
351
+ "mlflow.set_experiment(\"tourism-package-prediction\")\n",
352
+ "\n",
353
+ "api = HfApi()\n",
354
+ "\n",
355
+ "# Load the preprocessed data from Hugging Face\n",
356
+ "Xtrain_path = \"hf://datasets/ananttripathiak/tourism-dataset/Xtrain.csv\"\n",
357
+ "Xtest_path = \"hf://datasets/ananttripathiak/tourism-dataset/Xtest.csv\"\n",
358
+ "ytrain_path = \"hf://datasets/ananttripathiak/tourism-dataset/ytrain.csv\"\n",
359
+ "ytest_path = \"hf://datasets/ananttripathiak/tourism-dataset/ytest.csv\"\n",
360
+ "\n",
361
+ "print(\"Loading preprocessed data...\")\n",
362
+ "Xtrain = pd.read_csv(Xtrain_path)\n",
363
+ "Xtest = pd.read_csv(Xtest_path)\n",
364
+ "ytrain = pd.read_csv(ytrain_path).values.ravel()\n",
365
+ "ytest = pd.read_csv(ytest_path).values.ravel()\n",
366
+ "\n",
367
+ "print(f\"Training set shape: {Xtrain.shape}\")\n",
368
+ "print(f\"Test set shape: {Xtest.shape}\")\n",
369
+ "\n",
370
+ "# Identify numeric features (all features after encoding)\n",
371
+ "numeric_features = Xtrain.columns.tolist()\n",
372
+ "\n",
373
+ "# Preprocessor - StandardScaler for all numeric features\n",
374
+ "preprocessor = make_column_transformer(\n",
375
+ " (StandardScaler(), numeric_features)\n",
376
+ ")\n",
377
+ "\n",
378
+ "# Define base XGBoost Classifier\n",
379
+ "xgb_model = xgb.XGBClassifier(\n",
380
+ " random_state=42,\n",
381
+ " n_jobs=-1,\n",
382
+ " eval_metric='logloss',\n",
383
+ " use_label_encoder=False\n",
384
+ ")\n",
385
+ "\n",
386
+ "# Hyperparameter grid for classification\n",
387
+ "param_grid = {\n",
388
+ " 'xgbclassifier__n_estimators': [100, 200, 300],\n",
389
+ " 'xgbclassifier__max_depth': [3, 5, 7],\n",
390
+ " 'xgbclassifier__learning_rate': [0.01, 0.05, 0.1],\n",
391
+ " 'xgbclassifier__subsample': [0.7, 0.8, 1.0],\n",
392
+ " 'xgbclassifier__colsample_bytree': [0.7, 0.8, 1.0],\n",
393
+ " 'xgbclassifier__scale_pos_weight': [1, 2, 3] # Handle class imbalance\n",
394
+ "}\n",
395
+ "\n",
396
+ "# Pipeline\n",
397
+ "model_pipeline = make_pipeline(preprocessor, xgb_model)\n",
398
+ "\n",
399
+ "print(\"\\nStarting MLflow experiment...\")\n",
400
+ "with mlflow.start_run():\n",
401
+ " print(\"Performing Grid Search with Cross-Validation...\")\n",
402
+ " # Grid Search\n",
403
+ " grid_search = GridSearchCV(\n",
404
+ " model_pipeline,\n",
405
+ " param_grid,\n",
406
+ " cv=3,\n",
407
+ " n_jobs=-1,\n",
408
+ " scoring='roc_auc',\n",
409
+ " verbose=1\n",
410
+ " )\n",
411
+ " grid_search.fit(Xtrain, ytrain)\n",
412
+ "\n",
413
+ " # Log parameter sets\n",
414
+ " results = grid_search.cv_results_\n",
415
+ " print(f\"\\nEvaluated {len(results['params'])} parameter combinations\")\n",
416
+ "\n",
417
+ " for i in range(len(results['params'])):\n",
418
+ " param_set = results['params'][i]\n",
419
+ " mean_score = results['mean_test_score'][i]\n",
420
+ "\n",
421
+ " with mlflow.start_run(nested=True):\n",
422
+ " mlflow.log_params(param_set)\n",
423
+ " mlflow.log_metric(\"mean_roc_auc\", mean_score)\n",
424
+ "\n",
425
+ " # Best model\n",
426
+ " print(f\"\\nBest parameters: {grid_search.best_params_}\")\n",
427
+ " mlflow.log_params(grid_search.best_params_)\n",
428
+ " best_model = grid_search.best_estimator_\n",
429
+ "\n",
430
+ " # Predictions\n",
431
+ " print(\"\\nMaking predictions...\")\n",
432
+ " y_pred_train = best_model.predict(Xtrain)\n",
433
+ " y_pred_test = best_model.predict(Xtest)\n",
434
+ "\n",
435
+ " # Probability predictions for ROC-AUC\n",
436
+ " y_pred_train_proba = best_model.predict_proba(Xtrain)[:, 1]\n",
437
+ " y_pred_test_proba = best_model.predict_proba(Xtest)[:, 1]\n",
438
+ "\n",
439
+ " # Calculate metrics\n",
440
+ " print(\"\\nCalculating metrics...\")\n",
441
+ " train_accuracy = accuracy_score(ytrain, y_pred_train)\n",
442
+ " test_accuracy = accuracy_score(ytest, y_pred_test)\n",
443
+ "\n",
444
+ " train_precision = precision_score(ytrain, y_pred_train, zero_division=0)\n",
445
+ " test_precision = precision_score(ytest, y_pred_test, zero_division=0)\n",
446
+ "\n",
447
+ " train_recall = recall_score(ytrain, y_pred_train, zero_division=0)\n",
448
+ " test_recall = recall_score(ytest, y_pred_test, zero_division=0)\n",
449
+ "\n",
450
+ " train_f1 = f1_score(ytrain, y_pred_train, zero_division=0)\n",
451
+ " test_f1 = f1_score(ytest, y_pred_test, zero_division=0)\n",
452
+ "\n",
453
+ " train_roc_auc = roc_auc_score(ytrain, y_pred_train_proba)\n",
454
+ " test_roc_auc = roc_auc_score(ytest, y_pred_test_proba)\n",
455
+ "\n",
456
+ " # Log metrics\n",
457
+ " mlflow.log_metrics({\n",
458
+ " \"train_accuracy\": train_accuracy,\n",
459
+ " \"test_accuracy\": test_accuracy,\n",
460
+ " \"train_precision\": train_precision,\n",
461
+ " \"test_precision\": test_precision,\n",
462
+ " \"train_recall\": train_recall,\n",
463
+ " \"test_recall\": test_recall,\n",
464
+ " \"train_f1_score\": train_f1,\n",
465
+ " \"test_f1_score\": test_f1,\n",
466
+ " \"train_roc_auc\": train_roc_auc,\n",
467
+ " \"test_roc_auc\": test_roc_auc\n",
468
+ " })\n",
469
+ "\n",
470
+ " # Print results\n",
471
+ " print(\"\\n\" + \"=\"*50)\n",
472
+ " print(\"MODEL PERFORMANCE METRICS\")\n",
473
+ " print(\"=\"*50)\n",
474
+ " print(f\"Train Accuracy: {train_accuracy:.4f} | Test Accuracy: {test_accuracy:.4f}\")\n",
475
+ " print(f\"Train Precision: {train_precision:.4f} | Test Precision: {test_precision:.4f}\")\n",
476
+ " print(f\"Train Recall: {train_recall:.4f} | Test Recall: {test_recall:.4f}\")\n",
477
+ " print(f\"Train F1-Score: {train_f1:.4f} | Test F1-Score: {test_f1:.4f}\")\n",
478
+ " print(f\"Train ROC-AUC: {train_roc_auc:.4f} | Test ROC-AUC: {test_roc_auc:.4f}\")\n",
479
+ " print(\"=\"*50)\n",
480
+ "\n",
481
+ " print(\"\\nTest Set Classification Report:\")\n",
482
+ " print(classification_report(ytest, y_pred_test, target_names=['No Purchase', 'Purchase']))\n",
483
+ "\n",
484
+ " print(\"\\nTest Set Confusion Matrix:\")\n",
485
+ " print(confusion_matrix(ytest, y_pred_test))\n",
486
+ "\n",
487
+ " # Save the model locally\n",
488
+ " model_path = \"best_tourism_model_v1.joblib\"\n",
489
+ " joblib.dump(best_model, model_path)\n",
490
+ " print(f\"\\nModel saved locally as: {model_path}\")\n",
491
+ "\n",
492
+ " # Log the model artifact\n",
493
+ " mlflow.log_artifact(model_path, artifact_path=\"model\")\n",
494
+ " print(f\"Model logged to MLflow\")\n",
495
+ "\n",
496
+ " # Upload to Hugging Face\n",
497
+ " repo_id = \"ananttripathiak/tourism-prediction-model\"\n",
498
+ " repo_type = \"model\"\n",
499
+ "\n",
500
+ " # Step 1: Check if the repository exists\n",
501
+ " try:\n",
502
+ " api.repo_info(repo_id=repo_id, repo_type=repo_type)\n",
503
+ " print(f\"\\nRepository '{repo_id}' already exists. Using it.\")\n",
504
+ " except RepositoryNotFoundError:\n",
505
+ " print(f\"\\nRepository '{repo_id}' not found. Creating new repository...\")\n",
506
+ " create_repo(repo_id=repo_id, repo_type=repo_type, private=False)\n",
507
+ " print(f\"Repository '{repo_id}' created.\")\n",
508
+ "\n",
509
+ " # Upload model to Hugging Face\n",
510
+ " api.upload_file(\n",
511
+ " path_or_fileobj=\"best_tourism_model_v1.joblib\",\n",
512
+ " path_in_repo=\"best_tourism_model_v1.joblib\",\n",
513
+ " repo_id=repo_id,\n",
514
+ " repo_type=repo_type,\n",
515
+ " )\n",
516
+ " print(f\"Model uploaded to Hugging Face: {repo_id}\")\n",
517
+ "\n",
518
+ "print(\"\\n\" + \"=\"*50)\n",
519
+ "print(\"MODEL TRAINING COMPLETED SUCCESSFULLY!\")\n",
520
+ "print(\"=\"*50)"
521
+ ]
522
+ },
523
+ {
524
+ "cell_type": "code",
525
+ "execution_count": 7,
526
+ "metadata": {
527
+ "colab": {
528
+ "base_uri": "https://localhost:8080/"
529
+ },
530
+ "executionInfo": {
531
+ "elapsed": 203,
532
+ "status": "ok",
533
+ "timestamp": 1765705284390,
534
+ "user": {
535
+ "displayName": "Anant Tripathi",
536
+ "userId": "05588283814303116545"
537
+ },
538
+ "user_tz": -330
539
+ },
540
+ "id": "Vtjn_63uOtJ-",
541
+ "outputId": "259877f9-6be4-4ef1-cda4-c06d26720447"
542
+ },
543
+ "outputs": [
544
+ {
545
+ "name": "stdout",
546
+ "output_type": "stream",
547
+ "text": [
548
+ "Writing tourism_project/model_building/prep.py\n"
549
+ ]
550
+ }
551
+ ],
552
+ "source": [
553
+ "%%writefile tourism_project/model_building/prep.py\n",
554
+ "# for data manipulation\n",
555
+ "import pandas as pd\n",
556
+ "import numpy as np\n",
557
+ "# for data preprocessing and pipeline creation\n",
558
+ "from sklearn.model_selection import train_test_split\n",
559
+ "# for converting text data in to numerical representation\n",
560
+ "from sklearn.preprocessing import LabelEncoder\n",
561
+ "# for hugging face space authentication to upload files\n",
562
+ "from huggingface_hub import login, HfApi\n",
563
+ "import os\n",
564
+ "\n",
565
+ "# Define constants for the dataset and output paths\n",
566
+ "api = HfApi(token=os.getenv(\"HF_TOKEN\"))\n",
567
+ "DATASET_PATH = \"hf://datasets/ananttripathiak/tourism-dataset/tourism.csv\"\n",
568
+ "df = pd.read_csv(DATASET_PATH)\n",
569
+ "print(\"Dataset loaded successfully.\")\n",
570
+ "print(f\"Dataset shape: {df.shape}\")\n",
571
+ "\n",
572
+ "# Drop the unnamed index column if it exists\n",
573
+ "if 'Unnamed: 0' in df.columns or df.columns[0] == '':\n",
574
+ " df = df.iloc[:, 1:]\n",
575
+ "\n",
576
+ "# Drop CustomerID as it's a unique identifier (not useful for modeling)\n",
577
+ "if 'CustomerID' in df.columns:\n",
578
+ " df.drop(columns=['CustomerID'], inplace=True)\n",
579
+ "\n",
580
+ "# Handle missing values\n",
581
+ "print(\"\\nHandling missing values...\")\n",
582
+ "# For numerical columns, fill with median\n",
583
+ "numerical_cols = df.select_dtypes(include=[np.number]).columns\n",
584
+ "for col in numerical_cols:\n",
585
+ " if df[col].isnull().sum() > 0:\n",
586
+ " df[col].fillna(df[col].median(), inplace=True)\n",
587
+ "\n",
588
+ "# For categorical columns, fill with mode\n",
589
+ "categorical_cols = df.select_dtypes(include=['object']).columns\n",
590
+ "for col in categorical_cols:\n",
591
+ " if df[col].isnull().sum() > 0:\n",
592
+ " df[col].fillna(df[col].mode()[0], inplace=True)\n",
593
+ "\n",
594
+ "# Handle specific data quality issues (e.g., \"Fe Male\" should be \"Female\")\n",
595
+ "if 'Gender' in df.columns:\n",
596
+ " df['Gender'] = df['Gender'].str.strip().replace({'Fe Male': 'Female', 'Fe male': 'Female'})\n",
597
+ "\n",
598
+ "# Encode categorical columns\n",
599
+ "print(\"\\nEncoding categorical variables...\")\n",
600
+ "label_encoder = LabelEncoder()\n",
601
+ "\n",
602
+ "# List of categorical columns to encode\n",
603
+ "categorical_features = ['TypeofContact', 'Occupation', 'Gender', 'ProductPitched',\n",
604
+ " 'MaritalStatus', 'Designation']\n",
605
+ "\n",
606
+ "for col in categorical_features:\n",
607
+ " if col in df.columns:\n",
608
+ " df[col] = label_encoder.fit_transform(df[col].astype(str))\n",
609
+ "\n",
610
+ "# Define target variable\n",
611
+ "target_col = 'ProdTaken'\n",
612
+ "\n",
613
+ "# Split into X (features) and y (target)\n",
614
+ "X = df.drop(columns=[target_col])\n",
615
+ "y = df[target_col]\n",
616
+ "\n",
617
+ "print(f\"\\nFeatures shape: {X.shape}\")\n",
618
+ "print(f\"Target shape: {y.shape}\")\n",
619
+ "print(f\"Target distribution:\\n{y.value_counts()}\")\n",
620
+ "\n",
621
+ "# Perform train-test split\n",
622
+ "Xtrain, Xtest, ytrain, ytest = train_test_split(\n",
623
+ " X, y, test_size=0.2, random_state=42, stratify=y\n",
624
+ ")\n",
625
+ "\n",
626
+ "print(f\"\\nTrain set size: {Xtrain.shape[0]}\")\n",
627
+ "print(f\"Test set size: {Xtest.shape[0]}\")\n",
628
+ "\n",
629
+ "# Save the datasets\n",
630
+ "Xtrain.to_csv(\"Xtrain.csv\", index=False)\n",
631
+ "Xtest.to_csv(\"Xtest.csv\", index=False)\n",
632
+ "ytrain.to_csv(\"ytrain.csv\", index=False)\n",
633
+ "ytest.to_csv(\"ytest.csv\", index=False)\n",
634
+ "\n",
635
+ "print(\"\\nDatasets saved locally.\")\n",
636
+ "\n",
637
+ "# Upload to Hugging Face\n",
638
+ "files = [\"Xtrain.csv\", \"Xtest.csv\", \"ytrain.csv\", \"ytest.csv\"]\n",
639
+ "\n",
640
+ "for file_path in files:\n",
641
+ " api.upload_file(\n",
642
+ " path_or_fileobj=file_path,\n",
643
+ " path_in_repo=file_path.split(\"/\")[-1],\n",
644
+ " repo_id=\"ananttripathiak/tourism-dataset\",\n",
645
+ " repo_type=\"dataset\",\n",
646
+ " )\n",
647
+ " print(f\"Uploaded {file_path} to Hugging Face\")\n",
648
+ "\n",
649
+ "print(\"\\nData preparation completed successfully!\")"
650
+ ]
651
+ },
652
+ {
653
+ "cell_type": "markdown",
654
+ "metadata": {
655
+ "id": "0McYCZzkji5I"
656
+ },
657
+ "source": [
658
+ "# Deployment"
659
+ ]
660
+ },
661
+ {
662
+ "cell_type": "markdown",
663
+ "metadata": {
664
+ "id": "9QrY2v77vbEZ"
665
+ },
666
+ "source": [
667
+ "## Dockerfile"
668
+ ]
669
+ },
670
+ {
671
+ "cell_type": "code",
672
+ "execution_count": 8,
673
+ "metadata": {
674
+ "executionInfo": {
675
+ "elapsed": 2,
676
+ "status": "ok",
677
+ "timestamp": 1765705284393,
678
+ "user": {
679
+ "displayName": "Anant Tripathi",
680
+ "userId": "05588283814303116545"
681
+ },
682
+ "user_tz": -330
683
+ },
684
+ "id": "0-AMAI72CR-T"
685
+ },
686
+ "outputs": [],
687
+ "source": [
688
+ "os.makedirs(\"tourism_project/deployment\", exist_ok=True)"
689
+ ]
690
+ },
691
+ {
692
+ "cell_type": "code",
693
+ "execution_count": 9,
694
+ "metadata": {
695
+ "colab": {
696
+ "base_uri": "https://localhost:8080/"
697
+ },
698
+ "executionInfo": {
699
+ "elapsed": 10,
700
+ "status": "ok",
701
+ "timestamp": 1765705284404,
702
+ "user": {
703
+ "displayName": "Anant Tripathi",
704
+ "userId": "05588283814303116545"
705
+ },
706
+ "user_tz": -330
707
+ },
708
+ "id": "ZTicTDnPCVZr",
709
+ "outputId": "1707e165-9dee-4adb-c9c4-e2e0b50ce8fb"
710
+ },
711
+ "outputs": [
712
+ {
713
+ "name": "stdout",
714
+ "output_type": "stream",
715
+ "text": [
716
+ "Writing tourism_project/deployment/Dockerfile\n"
717
+ ]
718
+ }
719
+ ],
720
+ "source": [
721
+ "%%writefile tourism_project/deployment/Dockerfile\n",
722
+ "# Use a minimal base image with Python 3.9 installed\n",
723
+ "FROM python:3.9\n",
724
+ "\n",
725
+ "# Set the working directory inside the container to /app\n",
726
+ "WORKDIR /app\n",
727
+ "\n",
728
+ "# Copy all files from the current directory on the host to the container's /app directory\n",
729
+ "COPY . .\n",
730
+ "\n",
731
+ "# Install Python dependencies listed in requirements.txt\n",
732
+ "RUN pip3 install -r requirements.txt\n",
733
+ "\n",
734
+ "RUN useradd -m -u 1000 user\n",
735
+ "USER user\n",
736
+ "ENV HOME=/home/user \\\n",
737
+ "\tPATH=/home/user/.local/bin:$PATH\n",
738
+ "\n",
739
+ "WORKDIR $HOME/app\n",
740
+ "\n",
741
+ "COPY --chown=user . $HOME/app\n",
742
+ "\n",
743
+ "# Define the command to run the Streamlit app on port \"8501\" and make it accessible externally\n",
744
+ "CMD [\"streamlit\", \"run\", \"app.py\", \"--server.port=8501\", \"--server.address=0.0.0.0\", \"--server.enableXsrfProtection=false\"]"
745
+ ]
746
+ },
747
+ {
748
+ "cell_type": "markdown",
749
+ "metadata": {
750
+ "id": "LCvrklrBwNvJ"
751
+ },
752
+ "source": [
753
+ "## Streamlit App"
754
+ ]
755
+ },
756
+ {
757
+ "cell_type": "markdown",
758
+ "metadata": {
759
+ "id": "fXWe6ObRjP6-"
760
+ },
761
+ "source": [
762
+ "Please ensure that the web app script is named `app.py`."
763
+ ]
764
+ },
765
+ {
766
+ "cell_type": "code",
767
+ "execution_count": 10,
768
+ "metadata": {
769
+ "colab": {
770
+ "base_uri": "https://localhost:8080/"
771
+ },
772
+ "executionInfo": {
773
+ "elapsed": 56,
774
+ "status": "ok",
775
+ "timestamp": 1765705284459,
776
+ "user": {
777
+ "displayName": "Anant Tripathi",
778
+ "userId": "05588283814303116545"
779
+ },
780
+ "user_tz": -330
781
+ },
782
+ "id": "WBG-jxM89jdp",
783
+ "outputId": "885441f8-7cf2-4646-fc4e-705ebddf4640"
784
+ },
785
+ "outputs": [
786
+ {
787
+ "name": "stdout",
788
+ "output_type": "stream",
789
+ "text": [
790
+ "Writing tourism_project/deployment/app.py\n"
791
+ ]
792
+ }
793
+ ],
794
+ "source": [
795
+ "%%writefile tourism_project/deployment/app.py\n",
796
+ "import streamlit as st\n",
797
+ "import pandas as pd\n",
798
+ "import numpy as np\n",
799
+ "from huggingface_hub import hf_hub_download\n",
800
+ "import joblib\n",
801
+ "\n",
802
+ "# Page configuration\n",
803
+ "st.set_page_config(\n",
804
+ " page_title=\"Tourism Package Prediction\",\n",
805
+ " page_icon=\"✈️\",\n",
806
+ " layout=\"wide\"\n",
807
+ ")\n",
808
+ "\n",
809
+ "# Download and load the trained model\n",
810
+ "@st.cache_resource\n",
811
+ "def load_model():\n",
812
+ " try:\n",
813
+ " model_path = hf_hub_download(\n",
814
+ " repo_id=\"ananttripathiak/tourism-prediction-model\",\n",
815
+ " filename=\"best_tourism_model_v1.joblib\"\n",
816
+ " )\n",
817
+ " model = joblib.load(model_path)\n",
818
+ " return model\n",
819
+ " except Exception as e:\n",
820
+ " st.error(f\"Error loading model: {e}\")\n",
821
+ " return None\n",
822
+ "\n",
823
+ "model = load_model()\n",
824
+ "\n",
825
+ "# Title and Description\n",
826
+ "st.title(\"✈️ Wellness Tourism Package Prediction\")\n",
827
+ "st.markdown(\"\"\"\n",
828
+ "<style>\n",
829
+ " .main-header {\n",
830
+ " font-size: 20px;\n",
831
+ " color: #1f77b4;\n",
832
+ " margin-bottom: 20px;\n",
833
+ " }\n",
834
+ " .prediction-box {\n",
835
+ " padding: 20px;\n",
836
+ " border-radius: 10px;\n",
837
+ " margin: 20px 0;\n",
838
+ " }\n",
839
+ " .success-box {\n",
840
+ " background-color: #d4edda;\n",
841
+ " border: 1px solid #c3e6cb;\n",
842
+ " }\n",
843
+ " .warning-box {\n",
844
+ " background-color: #fff3cd;\n",
845
+ " border: 1px solid #ffeeba;\n",
846
+ " }\n",
847
+ "</style>\n",
848
+ "\"\"\", unsafe_allow_html=True)\n",
849
+ "\n",
850
+ "st.markdown(\"\"\"\n",
851
+ "This application predicts the likelihood of a customer purchasing the **Wellness Tourism Package**\n",
852
+ "based on their profile and interaction data. Enter customer details below to get a prediction.\n",
853
+ "\"\"\")\n",
854
+ "\n",
855
+ "# Create two columns for input\n",
856
+ "col1, col2 = st.columns(2)\n",
857
+ "\n",
858
+ "with col1:\n",
859
+ " st.subheader(\"📋 Customer Demographics\")\n",
860
+ "\n",
861
+ " age = st.number_input(\"Age\", min_value=18, max_value=100, value=35, step=1)\n",
862
+ "\n",
863
+ " type_of_contact = st.selectbox(\n",
864
+ " \"Type of Contact\",\n",
865
+ " options=[\"Self Enquiry\", \"Company Invited\"]\n",
866
+ " )\n",
867
+ "\n",
868
+ " city_tier = st.selectbox(\n",
869
+ " \"City Tier\",\n",
870
+ " options=[1, 2, 3],\n",
871
+ " help=\"Tier 1: Metro cities, Tier 2: Mid-sized cities, Tier 3: Smaller cities\"\n",
872
+ " )\n",
873
+ "\n",
874
+ " occupation = st.selectbox(\n",
875
+ " \"Occupation\",\n",
876
+ " options=[\"Salaried\", \"Small Business\", \"Free Lancer\", \"Large Business\"]\n",
877
+ " )\n",
878
+ "\n",
879
+ " gender = st.selectbox(\"Gender\", options=[\"Male\", \"Female\"])\n",
880
+ "\n",
881
+ " marital_status = st.selectbox(\n",
882
+ " \"Marital Status\",\n",
883
+ " options=[\"Single\", \"Married\", \"Divorced\", \"Unmarried\"]\n",
884
+ " )\n",
885
+ "\n",
886
+ " designation = st.selectbox(\n",
887
+ " \"Designation\",\n",
888
+ " options=[\"Executive\", \"Manager\", \"Senior Manager\", \"AVP\", \"VP\"]\n",
889
+ " )\n",
890
+ "\n",
891
+ " monthly_income = st.number_input(\n",
892
+ " \"Monthly Income (₹)\",\n",
893
+ " min_value=0.0,\n",
894
+ " max_value=200000.0,\n",
895
+ " value=25000.0,\n",
896
+ " step=1000.0\n",
897
+ " )\n",
898
+ "\n",
899
+ "with col2:\n",
900
+ " st.subheader(\"🎯 Customer Interaction & Preferences\")\n",
901
+ "\n",
902
+ " duration_of_pitch = st.number_input(\n",
903
+ " \"Duration of Pitch (minutes)\",\n",
904
+ " min_value=0.0,\n",
905
+ " max_value=60.0,\n",
906
+ " value=15.0,\n",
907
+ " step=0.5\n",
908
+ " )\n",
909
+ "\n",
910
+ " number_of_persons_visiting = st.number_input(\n",
911
+ " \"Number of Persons Visiting\",\n",
912
+ " min_value=1,\n",
913
+ " max_value=10,\n",
914
+ " value=2,\n",
915
+ " step=1\n",
916
+ " )\n",
917
+ "\n",
918
+ " number_of_followups = st.number_input(\n",
919
+ " \"Number of Follow-ups\",\n",
920
+ " min_value=0.0,\n",
921
+ " max_value=10.0,\n",
922
+ " value=3.0,\n",
923
+ " step=1.0\n",
924
+ " )\n",
925
+ "\n",
926
+ " product_pitched = st.selectbox(\n",
927
+ " \"Product Pitched\",\n",
928
+ " options=[\"Basic\", \"Standard\", \"Deluxe\", \"Super Deluxe\", \"King\"]\n",
929
+ " )\n",
930
+ "\n",
931
+ " preferred_property_star = st.selectbox(\n",
932
+ " \"Preferred Property Star Rating\",\n",
933
+ " options=[3.0, 4.0, 5.0]\n",
934
+ " )\n",
935
+ "\n",
936
+ " number_of_trips = st.number_input(\n",
937
+ " \"Number of Trips (per year)\",\n",
938
+ " min_value=0.0,\n",
939
+ " max_value=20.0,\n",
940
+ " value=3.0,\n",
941
+ " step=1.0\n",
942
+ " )\n",
943
+ "\n",
944
+ " passport = st.selectbox(\"Has Passport?\", options=[\"Yes\", \"No\"])\n",
945
+ "\n",
946
+ " pitch_satisfaction_score = st.slider(\n",
947
+ " \"Pitch Satisfaction Score\",\n",
948
+ " min_value=1,\n",
949
+ " max_value=5,\n",
950
+ " value=3,\n",
951
+ " step=1\n",
952
+ " )\n",
953
+ "\n",
954
+ " own_car = st.selectbox(\"Owns Car?\", options=[\"Yes\", \"No\"])\n",
955
+ "\n",
956
+ " number_of_children_visiting = st.number_input(\n",
957
+ " \"Number of Children Visiting\",\n",
958
+ " min_value=0.0,\n",
959
+ " max_value=5.0,\n",
960
+ " value=0.0,\n",
961
+ " step=1.0\n",
962
+ " )\n",
963
+ "\n",
964
+ "# Encoding mapping (based on LabelEncoder used during training)\n",
965
+ "# Note: These mappings should match the exact encoding used during training\n",
966
+ "type_of_contact_map = {\"Company Invited\": 0, \"Self Enquiry\": 1}\n",
967
+ "occupation_map = {\"Free Lancer\": 0, \"Large Business\": 1, \"Salaried\": 2, \"Small Business\": 3}\n",
968
+ "gender_map = {\"Female\": 0, \"Male\": 1}\n",
969
+ "product_pitched_map = {\"Basic\": 0, \"Deluxe\": 1, \"King\": 2, \"Standard\": 3, \"Super Deluxe\": 4}\n",
970
+ "marital_status_map = {\"Divorced\": 0, \"Married\": 1, \"Single\": 2, \"Unmarried\": 3}\n",
971
+ "designation_map = {\"AVP\": 0, \"Executive\": 1, \"Manager\": 2, \"Senior Manager\": 3, \"VP\": 4}\n",
972
+ "\n",
973
+ "# Convert Yes/No to 0/1\n",
974
+ "passport_val = 1 if passport == \"Yes\" else 0\n",
975
+ "own_car_val = 1 if own_car == \"Yes\" else 0\n",
976
+ "\n",
977
+ "# Prepare input data\n",
978
+ "input_data = pd.DataFrame([{\n",
979
+ " 'Age': age,\n",
980
+ " 'TypeofContact': type_of_contact_map[type_of_contact],\n",
981
+ " 'CityTier': city_tier,\n",
982
+ " 'DurationOfPitch': duration_of_pitch,\n",
983
+ " 'Occupation': occupation_map[occupation],\n",
984
+ " 'Gender': gender_map[gender],\n",
985
+ " 'NumberOfPersonVisiting': number_of_persons_visiting,\n",
986
+ " 'NumberOfFollowups': number_of_followups,\n",
987
+ " 'ProductPitched': product_pitched_map[product_pitched],\n",
988
+ " 'PreferredPropertyStar': preferred_property_star,\n",
989
+ " 'MaritalStatus': marital_status_map[marital_status],\n",
990
+ " 'NumberOfTrips': number_of_trips,\n",
991
+ " 'Passport': passport_val,\n",
992
+ " 'PitchSatisfactionScore': pitch_satisfaction_score,\n",
993
+ " 'OwnCar': own_car_val,\n",
994
+ " 'NumberOfChildrenVisiting': number_of_children_visiting,\n",
995
+ " 'Designation': designation_map[designation],\n",
996
+ " 'MonthlyIncome': monthly_income\n",
997
+ "}])\n",
998
+ "\n",
999
+ "# Predict button\n",
1000
+ "st.markdown(\"---\")\n",
1001
+ "if st.button(\"🔮 Predict Purchase Likelihood\", use_container_width=True):\n",
1002
+ " if model is not None:\n",
1003
+ " try:\n",
1004
+ " # Make prediction\n",
1005
+ " prediction = model.predict(input_data)[0]\n",
1006
+ " prediction_proba = model.predict_proba(input_data)[0]\n",
1007
+ "\n",
1008
+ " # Display results\n",
1009
+ " st.markdown(\"### 📊 Prediction Results\")\n",
1010
+ "\n",
1011
+ " if prediction == 1:\n",
1012
+ " st.markdown(\n",
1013
+ " f'<div class=\"prediction-box success-box\">'\n",
1014
+ " f'<h2 style=\"color: #155724;\">✅ High Likelihood of Purchase!</h2>'\n",
1015
+ " f'<p style=\"font-size: 18px;\">This customer is <b>likely to purchase</b> the Wellness Tourism Package.</p>'\n",
1016
+ " f'<p style=\"font-size: 16px;\">Confidence: <b>{prediction_proba[1]*100:.2f}%</b></p>'\n",
1017
+ " f'</div>',\n",
1018
+ " unsafe_allow_html=True\n",
1019
+ " )\n",
1020
+ " st.success(\"💡 **Recommendation:** Prioritize follow-up with this customer!\")\n",
1021
+ " else:\n",
1022
+ " st.markdown(\n",
1023
+ " f'<div class=\"prediction-box warning-box\">'\n",
1024
+ " f'<h2 style=\"color: #856404;\">⚠️ Low Likelihood of Purchase</h2>'\n",
1025
+ " f'<p style=\"font-size: 18px;\">This customer is <b>unlikely to purchase</b> the Wellness Tourism Package.</p>'\n",
1026
+ " f'<p style=\"font-size: 16px;\">Confidence: <b>{prediction_proba[0]*100:.2f}%</b></p>'\n",
1027
+ " f'</div>',\n",
1028
+ " unsafe_allow_html=True\n",
1029
+ " )\n",
1030
+ " st.info(\"💡 **Recommendation:** Consider alternative packages or additional engagement strategies.\")\n",
1031
+ "\n",
1032
+ " # Show probability breakdown\n",
1033
+ " col_prob1, col_prob2 = st.columns(2)\n",
1034
+ " with col_prob1:\n",
1035
+ " st.metric(\"Probability of No Purchase\", f\"{prediction_proba[0]*100:.2f}%\")\n",
1036
+ " with col_prob2:\n",
1037
+ " st.metric(\"Probability of Purchase\", f\"{prediction_proba[1]*100:.2f}%\")\n",
1038
+ "\n",
1039
+ " except Exception as e:\n",
1040
+ " st.error(f\"Error making prediction: {e}\")\n",
1041
+ " else:\n",
1042
+ " st.error(\"Model not loaded. Please check the model repository.\")\n",
1043
+ "\n",
1044
+ "# Footer\n",
1045
+ "st.markdown(\"---\")\n",
1046
+ "st.markdown(\"\"\"\n",
1047
+ "<div style='text-align: center; color: #666; padding: 20px;'>\n",
1048
+ " <p>🏢 Visit with Us - Wellness Tourism Package Prediction System</p>\n",
1049
+ " <p>Built with ❤️ using Streamlit and XGBoost</p>\n",
1050
+ "</div>\n",
1051
+ "\"\"\", unsafe_allow_html=True)"
1052
+ ]
1053
+ },
1054
+ {
1055
+ "cell_type": "markdown",
1056
+ "metadata": {
1057
+ "id": "07cYzWcIwTL-"
1058
+ },
1059
+ "source": [
1060
+ "## Dependency Handling"
1061
+ ]
1062
+ },
1063
+ {
1064
+ "cell_type": "markdown",
1065
+ "metadata": {
1066
+ "id": "JEgfHL64jU7o"
1067
+ },
1068
+ "source": [
1069
+ "Please ensure that the dependency handling file is named `requirements.txt`."
1070
+ ]
1071
+ },
1072
+ {
1073
+ "cell_type": "code",
1074
+ "execution_count": 11,
1075
+ "metadata": {
1076
+ "colab": {
1077
+ "base_uri": "https://localhost:8080/"
1078
+ },
1079
+ "executionInfo": {
1080
+ "elapsed": 4,
1081
+ "status": "ok",
1082
+ "timestamp": 1765705284460,
1083
+ "user": {
1084
+ "displayName": "Anant Tripathi",
1085
+ "userId": "05588283814303116545"
1086
+ },
1087
+ "user_tz": -330
1088
+ },
1089
+ "id": "nvdmy7Wd9lda",
1090
+ "outputId": "16b57c41-8e07-49cc-d6c5-1f15240b59fe"
1091
+ },
1092
+ "outputs": [
1093
+ {
1094
+ "name": "stdout",
1095
+ "output_type": "stream",
1096
+ "text": [
1097
+ "Writing tourism_project/deployment/requirements.txt\n"
1098
+ ]
1099
+ }
1100
+ ],
1101
+ "source": [
1102
+ "%%writefile tourism_project/deployment/requirements.txt\n",
1103
+ "pandas==2.2.2\n",
1104
+ "numpy==1.26.4\n",
1105
+ "huggingface_hub==0.32.6\n",
1106
+ "streamlit==1.43.2\n",
1107
+ "joblib==1.5.1\n",
1108
+ "scikit-learn==1.6.0\n",
1109
+ "xgboost==2.1.4"
1110
+ ]
1111
+ },
1112
+ {
1113
+ "cell_type": "markdown",
1114
+ "metadata": {
1115
+ "id": "V4ynzpKNwWS_"
1116
+ },
1117
+ "source": [
1118
+ "# Hosting"
1119
+ ]
1120
+ },
1121
+ {
1122
+ "cell_type": "code",
1123
+ "execution_count": 12,
1124
+ "metadata": {
1125
+ "executionInfo": {
1126
+ "elapsed": 1,
1127
+ "status": "ok",
1128
+ "timestamp": 1765705284460,
1129
+ "user": {
1130
+ "displayName": "Anant Tripathi",
1131
+ "userId": "05588283814303116545"
1132
+ },
1133
+ "user_tz": -330
1134
+ },
1135
+ "id": "7p5sBvTg9nCW"
1136
+ },
1137
+ "outputs": [],
1138
+ "source": [
1139
+ "os.makedirs(\"tourism_project/hosting\", exist_ok=True)"
1140
+ ]
1141
+ },
1142
+ {
1143
+ "cell_type": "code",
1144
+ "execution_count": 13,
1145
+ "metadata": {
1146
+ "colab": {
1147
+ "base_uri": "https://localhost:8080/"
1148
+ },
1149
+ "executionInfo": {
1150
+ "elapsed": 2,
1151
+ "status": "ok",
1152
+ "timestamp": 1765705284463,
1153
+ "user": {
1154
+ "displayName": "Anant Tripathi",
1155
+ "userId": "05588283814303116545"
1156
+ },
1157
+ "user_tz": -330
1158
+ },
1159
+ "id": "QlpgAQoXOtKA",
1160
+ "outputId": "2611020e-c3cf-4dea-af77-ccb122c85e65"
1161
+ },
1162
+ "outputs": [
1163
+ {
1164
+ "name": "stdout",
1165
+ "output_type": "stream",
1166
+ "text": [
1167
+ "Writing tourism_project/hosting/hosting.py\n"
1168
+ ]
1169
+ }
1170
+ ],
1171
+ "source": [
1172
+ "%%writefile tourism_project/hosting/hosting.py\n",
1173
+ "from huggingface_hub import HfApi\n",
1174
+ "import os\n",
1175
+ "\n",
1176
+ "api = HfApi(token=os.getenv(\"HF_TOKEN\"))\n",
1177
+ "api.upload_folder(\n",
1178
+ " folder_path=\"tourism_project/deployment\",\n",
1179
+ " repo_id=\"ananttripathiak/wellness-tourism-prediction\",\n",
1180
+ " repo_type=\"space\",\n",
1181
+ " path_in_repo=\"\",\n",
1182
+ ")\n",
1183
+ "\n",
1184
+ "print(\"Deployment files successfully uploaded to Hugging Face Space!\")"
1185
+ ]
1186
+ },
1187
+ {
1188
+ "cell_type": "markdown",
1189
+ "metadata": {
1190
+ "id": "PuCgAW2hktli"
1191
+ },
1192
+ "source": [
1193
+ "# MLOps Pipeline with Github Actions Workflow"
1194
+ ]
1195
+ },
1196
+ {
1197
+ "cell_type": "markdown",
1198
+ "metadata": {
1199
+ "id": "L5BZr5i8PKVN"
1200
+ },
1201
+ "source": [
1202
+ "**Note:**\n",
1203
+ "\n",
1204
+ "1. Before running the file below, make sure to add the HF_TOKEN to your GitHub secrets to enable authentication between GitHub and Hugging Face.\n",
1205
+ "2. The below code is for a sample YAML file that can be updated as required to meet the requirements of this project."
1206
+ ]
1207
+ },
1208
+ {
1209
+ "cell_type": "code",
1210
+ "execution_count": 14,
1211
+ "metadata": {
1212
+ "executionInfo": {
1213
+ "elapsed": 10,
1214
+ "status": "ok",
1215
+ "timestamp": 1765705284474,
1216
+ "user": {
1217
+ "displayName": "Anant Tripathi",
1218
+ "userId": "05588283814303116545"
1219
+ },
1220
+ "user_tz": -330
1221
+ },
1222
+ "id": "H2kXyghROtKA"
1223
+ },
1224
+ "outputs": [],
1225
+ "source": [
1226
+ "os.makedirs(\"tourism_project/.github/workflows\", exist_ok=True)"
1227
+ ]
1228
+ },
1229
+ {
1230
+ "cell_type": "code",
1231
+ "execution_count": 15,
1232
+ "metadata": {
1233
+ "colab": {
1234
+ "base_uri": "https://localhost:8080/"
1235
+ },
1236
+ "executionInfo": {
1237
+ "elapsed": 5,
1238
+ "status": "ok",
1239
+ "timestamp": 1765705284479,
1240
+ "user": {
1241
+ "displayName": "Anant Tripathi",
1242
+ "userId": "05588283814303116545"
1243
+ },
1244
+ "user_tz": -330
1245
+ },
1246
+ "id": "M5J4Kq2ROtKA",
1247
+ "outputId": "a9c5ad71-d5b5-41a0-823e-8739942162fb"
1248
+ },
1249
+ "outputs": [
1250
+ {
1251
+ "name": "stdout",
1252
+ "output_type": "stream",
1253
+ "text": [
1254
+ "Writing tourism_project/.github/workflows/pipeline.yml\n"
1255
+ ]
1256
+ }
1257
+ ],
1258
+ "source": [
1259
+ "%%writefile tourism_project/.github/workflows/pipeline.yml\n",
1260
+ "name: Tourism Package Prediction Pipeline\n",
1261
+ "\n",
1262
+ "on:\n",
1263
+ " push:\n",
1264
+ " branches:\n",
1265
+ " - main\n",
1266
+ "\n",
1267
+ "jobs:\n",
1268
+ "\n",
1269
+ " register-dataset:\n",
1270
+ " runs-on: ubuntu-latest\n",
1271
+ " steps:\n",
1272
+ " - uses: actions/checkout@v3\n",
1273
+ " - name: Install Dependencies\n",
1274
+ " run: pip install -r tourism_project/requirements.txt\n",
1275
+ " - name: Upload Dataset to Hugging Face Hub\n",
1276
+ " env:\n",
1277
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1278
+ " run: python tourism_project/model_building/data_register.py\n",
1279
+ "\n",
1280
+ " data-prep:\n",
1281
+ " needs: register-dataset\n",
1282
+ " runs-on: ubuntu-latest\n",
1283
+ " steps:\n",
1284
+ " - uses: actions/checkout@v3\n",
1285
+ " - name: Install Dependencies\n",
1286
+ " run: pip install -r tourism_project/requirements.txt\n",
1287
+ " - name: Run Data Preparation\n",
1288
+ " env:\n",
1289
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1290
+ " run: python tourism_project/model_building/prep.py\n",
1291
+ "\n",
1292
+ " model-traning:\n",
1293
+ " needs: data-prep\n",
1294
+ " runs-on: ubuntu-latest\n",
1295
+ " steps:\n",
1296
+ " - uses: actions/checkout@v3\n",
1297
+ " - name: Install Dependencies\n",
1298
+ " run: pip install -r tourism_project/requirements.txt\n",
1299
+ " - name: Start MLflow Server\n",
1300
+ " run: |\n",
1301
+ " nohup mlflow ui --host 0.0.0.0 --port 5000 &\n",
1302
+ " sleep 5\n",
1303
+ " - name: Model Building\n",
1304
+ " env:\n",
1305
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1306
+ " run: python tourism_project/model_building/train.py\n",
1307
+ "\n",
1308
+ " deploy-hosting:\n",
1309
+ " runs-on: ubuntu-latest\n",
1310
+ " needs: [model-traning,data-prep,register-dataset]\n",
1311
+ " steps:\n",
1312
+ " - uses: actions/checkout@v3\n",
1313
+ " - name: Install Dependencies\n",
1314
+ " run: pip install -r tourism_project/requirements.txt\n",
1315
+ " - name: Push files to Frontend Hugging Face Space\n",
1316
+ " env:\n",
1317
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1318
+ " run: python tourism_project/hosting/hosting.py\n"
1319
+ ]
1320
+ },
1321
+ {
1322
+ "cell_type": "markdown",
1323
+ "metadata": {
1324
+ "id": "J029tYPq4Rmq"
1325
+ },
1326
+ "source": [
1327
+ "```\n",
1328
+ "name: Tourism Project Pipeline\n",
1329
+ "\n",
1330
+ "on:\n",
1331
+ " push:\n",
1332
+ " branches:\n",
1333
+ " - main # Automatically triggers on push to the main branch\n",
1334
+ "\n",
1335
+ "jobs:\n",
1336
+ "\n",
1337
+ " register-dataset:\n",
1338
+ " runs-on: ubuntu-latest\n",
1339
+ " steps:\n",
1340
+ " - uses: actions/checkout@v3\n",
1341
+ " - name: Install Dependencies\n",
1342
+ " run: <add_code_here>\n",
1343
+ " - name: Upload Dataset to Hugging Face Hub\n",
1344
+ " env:\n",
1345
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1346
+ " run: <add_code_here>\n",
1347
+ "\n",
1348
+ " data-prep:\n",
1349
+ " needs: register-dataset\n",
1350
+ " runs-on: ubuntu-latest\n",
1351
+ " steps:\n",
1352
+ " - uses: actions/checkout@v3\n",
1353
+ " - name: Install Dependencies\n",
1354
+ " run: <add_code_here>\n",
1355
+ " - name: Run Data Preparation\n",
1356
+ " env:\n",
1357
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1358
+ " run: <add_code_here>\n",
1359
+ "\n",
1360
+ "\n",
1361
+ " model-traning:\n",
1362
+ " needs: data-prep\n",
1363
+ " runs-on: ubuntu-latest\n",
1364
+ " steps:\n",
1365
+ " - uses: actions/checkout@v3\n",
1366
+ " - name: Install Dependencies\n",
1367
+ " run: <add_code_here>\n",
1368
+ " - name: Start MLflow Server\n",
1369
+ " run: |\n",
1370
+ " nohup mlflow ui --host 0.0.0.0 --port 5000 & # Run MLflow UI in the background\n",
1371
+ " sleep 5 # Wait for a moment to let the server starts\n",
1372
+ " - name: Model Building\n",
1373
+ " env:\n",
1374
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1375
+ " run: <add_code_here>\n",
1376
+ "\n",
1377
+ "\n",
1378
+ " deploy-hosting:\n",
1379
+ " runs-on: ubuntu-latest\n",
1380
+ " needs: [model-traning,data-prep,register-dataset]\n",
1381
+ " steps:\n",
1382
+ " - uses: actions/checkout@v3\n",
1383
+ " - name: Install Dependencies\n",
1384
+ " run: <add_code_here>\n",
1385
+ " - name: Push files to Frontend Hugging Face Space\n",
1386
+ " env:\n",
1387
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1388
+ " run: <add_code_here>\n",
1389
+ "\n",
1390
+ "```"
1391
+ ]
1392
+ },
1393
+ {
1394
+ "cell_type": "markdown",
1395
+ "metadata": {
1396
+ "id": "T9fgZ_Mq3zzp"
1397
+ },
1398
+ "source": [
1399
+ "**Note:** To use this YAML file for our use case, we need to\n",
1400
+ "\n",
1401
+ "1. Go to the GitHub repository for the project\n",
1402
+ "2. Create a folder named ***.github/workflows/***\n",
1403
+ "3. In the above folder, create a file named ***pipeline.yml***\n",
1404
+ "4. Copy and paste the above content for the YAML file into the ***pipeline.yml*** file"
1405
+ ]
1406
+ },
1407
+ {
1408
+ "cell_type": "markdown",
1409
+ "metadata": {
1410
+ "id": "PvEUJ-t5kdxH"
1411
+ },
1412
+ "source": [
1413
+ "## Requirements file for the Github Actions Workflow"
1414
+ ]
1415
+ },
1416
+ {
1417
+ "cell_type": "code",
1418
+ "execution_count": 16,
1419
+ "metadata": {
1420
+ "colab": {
1421
+ "base_uri": "https://localhost:8080/"
1422
+ },
1423
+ "executionInfo": {
1424
+ "elapsed": 2,
1425
+ "status": "ok",
1426
+ "timestamp": 1765705284481,
1427
+ "user": {
1428
+ "displayName": "Anant Tripathi",
1429
+ "userId": "05588283814303116545"
1430
+ },
1431
+ "user_tz": -330
1432
+ },
1433
+ "id": "nfqWcLRm-dga",
1434
+ "outputId": "ca9d9001-8154-47f1-809a-c2d62e51b4c2"
1435
+ },
1436
+ "outputs": [
1437
+ {
1438
+ "name": "stdout",
1439
+ "output_type": "stream",
1440
+ "text": [
1441
+ "Writing tourism_project/requirements.txt\n"
1442
+ ]
1443
+ }
1444
+ ],
1445
+ "source": [
1446
+ "%%writefile tourism_project/requirements.txt\n",
1447
+ "huggingface_hub==0.32.6\n",
1448
+ "datasets==3.6.0\n",
1449
+ "pandas==2.2.2\n",
1450
+ "numpy==1.26.4\n",
1451
+ "scikit-learn==1.6.0\n",
1452
+ "xgboost==2.1.4\n",
1453
+ "mlflow==3.0.1\n",
1454
+ "joblib==1.5.1"
1455
+ ]
1456
+ },
1457
+ {
1458
+ "cell_type": "markdown",
1459
+ "metadata": {
1460
+ "id": "BA6mP-Ebkm3O"
1461
+ },
1462
+ "source": [
1463
+ "## Github Authentication and Push Files"
1464
+ ]
1465
+ },
1466
+ {
1467
+ "cell_type": "markdown",
1468
+ "metadata": {
1469
+ "id": "T84Ei-g9Z2uw"
1470
+ },
1471
+ "source": [
1472
+ "* Before moving forward, we need to generate a secret token to push files directly from Colab to the GitHub repository.\n",
1473
+ "* Please follow the below instructions to create the GitHub token:\n",
1474
+ " - Open your GitHub profile.\n",
1475
+ " - Click on ***Settings***.\n",
1476
+ " - Go to ***Developer Settings***.\n",
1477
+ " - Expand the ***Personal access tokens*** section and select ***Tokens (classic)***.\n",
1478
+ " - Click ***Generate new token***, then choose ***Generate new token (classic)***.\n",
1479
+ " - Add a note and select all required scopes.\n",
1480
+ " - Click ***Generate token***.\n",
1481
+ " - Copy the generated token and store it safely in a notepad."
1482
+ ]
1483
+ },
1484
+ {
1485
+ "cell_type": "code",
1486
+ "execution_count": 17,
1487
+ "metadata": {
1488
+ "colab": {
1489
+ "base_uri": "https://localhost:8080/"
1490
+ },
1491
+ "executionInfo": {
1492
+ "elapsed": 6682,
1493
+ "status": "ok",
1494
+ "timestamp": 1765705291163,
1495
+ "user": {
1496
+ "displayName": "Anant Tripathi",
1497
+ "userId": "05588283814303116545"
1498
+ },
1499
+ "user_tz": -330
1500
+ },
1501
+ "id": "KPDx4gqGh7cO",
1502
+ "outputId": "47eb6218-6a20-4eae-ee1d-04c771d14101"
1503
+ },
1504
+ "outputs": [
1505
+ {
1506
+ "name": "stdout",
1507
+ "output_type": "stream",
1508
+ "text": [
1509
+ "Reading package lists... Done\n",
1510
+ "Building dependency tree... Done\n",
1511
+ "Reading state information... Done\n",
1512
+ "git is already the newest version (1:2.34.1-1ubuntu1.15).\n",
1513
+ "0 upgraded, 0 newly installed, 0 to remove and 1 not upgraded.\n",
1514
+ "Cloning into 'Tourism_Project'...\n",
1515
+ "remote: Enumerating objects: 99, done.\u001b[K\n",
1516
+ "remote: Counting objects: 100% (99/99), done.\u001b[K\n",
1517
+ "remote: Compressing objects: 100% (89/89), done.\u001b[K\n",
1518
+ "remote: Total 99 (delta 31), reused 0 (delta 0), pack-reused 0 (from 0)\u001b[K\n",
1519
+ "Receiving objects: 100% (99/99), 111.21 KiB | 4.28 MiB/s, done.\n",
1520
+ "Resolving deltas: 100% (31/31), done.\n",
1521
+ "mv: cannot move '/content/tourism_project/' to '/content/Tourism_Project/tourism_project': Directory not empty\n"
1522
+ ]
1523
+ }
1524
+ ],
1525
+ "source": [
1526
+ "# Install Git\n",
1527
+ "!apt-get install git\n",
1528
+ "\n",
1529
+ "# Set your Git identity (replace with your details)\n",
1530
+ "!git config --global user.email \"ananttripathi1996@gmail.com\"\n",
1531
+ "!git config --global user.name \"ananttripathi\"\n",
1532
+ "\n",
1533
+ "# Clone your GitHub repository\n",
1534
+ "!git clone https://github.com/ananttripathi/Tourism_Project.git\n",
1535
+ "\n",
1536
+ "# Move your folder to the repository directory\n",
1537
+ "!mv /content/tourism_project/ /content/Tourism_Project"
1538
+ ]
1539
+ },
1540
+ {
1541
+ "cell_type": "code",
1542
+ "execution_count": null,
1543
+ "metadata": {
1544
+ "colab": {
1545
+ "base_uri": "https://localhost:8080/"
1546
+ },
1547
+ "executionInfo": {
1548
+ "elapsed": 435,
1549
+ "status": "ok",
1550
+ "timestamp": 1765705291611,
1551
+ "user": {
1552
+ "displayName": "Anant Tripathi",
1553
+ "userId": "05588283814303116545"
1554
+ },
1555
+ "user_tz": -330
1556
+ },
1557
+ "id": "IuUahCwVigon",
1558
+ "outputId": "4ca2922a-9c3b-4bbe-ce2f-ef55236d9ba5"
1559
+ },
1560
+ "outputs": [
1561
+ {
1562
+ "name": "stdout",
1563
+ "output_type": "stream",
1564
+ "text": [
1565
+ "/content/Tourism_Project\n",
1566
+ "On branch main\n",
1567
+ "Your branch is up to date with 'origin/main'.\n",
1568
+ "\n",
1569
+ "nothing to commit, working tree clean\n",
1570
+ "Everything up-to-date\n"
1571
+ ]
1572
+ }
1573
+ ],
1574
+ "source": [
1575
+ "# Change directory to the cloned repository\n",
1576
+ "%cd Tourism_Project/\n",
1577
+ "\n",
1578
+ "# Add the new folder to Git\n",
1579
+ "!git add .\n",
1580
+ "\n",
1581
+ "# Commit the changes\n",
1582
+ "!git commit -m \"first commit\"\n",
1583
+ "\n",
1584
+ "# Push to GitHub (you'll need your GitHub credentials; use a personal access token if 2FA enabled)\n",
1585
+ "!git push https://ananttripathi:github_pat_11SPxp30Fxq@github.com/ananttripathi/Tourism_Project.git"
1586
+ ]
1587
+ },
1588
+ {
1589
+ "cell_type": "markdown",
1590
+ "metadata": {
1591
+ "id": "v-i8Jdyz-_L1"
1592
+ },
1593
+ "source": [
1594
+ "# Output Evaluation"
1595
+ ]
1596
+ },
1597
+ {
1598
+ "cell_type": "markdown",
1599
+ "metadata": {
1600
+ "id": "FTK8Bpda_UHg"
1601
+ },
1602
+ "source": [
1603
+ "- GitHub (link to repository, screenshot of folder structure and executed workflow)"
1604
+ ]
1605
+ },
1606
+ {
1607
+ "cell_type": "code",
1608
+ "execution_count": 19,
1609
+ "metadata": {
1610
+ "executionInfo": {
1611
+ "elapsed": 57,
1612
+ "status": "ok",
1613
+ "timestamp": 1765705291669,
1614
+ "user": {
1615
+ "displayName": "Anant Tripathi",
1616
+ "userId": "05588283814303116545"
1617
+ },
1618
+ "user_tz": -330
1619
+ },
1620
+ "id": "6qzzesaG_Xw8"
1621
+ },
1622
+ "outputs": [],
1623
+ "source": [
1624
+ "# TODO: Add Screenshot After Deployment\n",
1625
+ "#\n",
1626
+ "# After pushing to GitHub and running the workflow, add a screenshot showing:\n",
1627
+ "# 1. GitHub repository structure (showing tourism_project folder and all subfolders)\n",
1628
+ "# 2. GitHub Actions workflow execution (all 4 jobs completed successfully with green checkmarks)\n",
1629
+ "#\n",
1630
+ "# Screenshot should show:\n",
1631
+ "# ✅ register-dataset job completed\n",
1632
+ "# ✅ data-prep job completed\n",
1633
+ "# ✅ model-training job completed\n",
1634
+ "# ✅ deploy-hosting job completed\n",
1635
+ "#\n",
1636
+ "# You can also provide the GitHub repository URL here"
1637
+ ]
1638
+ },
1639
+ {
1640
+ "cell_type": "code",
1641
+ "execution_count": 19,
1642
+ "metadata": {
1643
+ "executionInfo": {
1644
+ "elapsed": 2,
1645
+ "status": "ok",
1646
+ "timestamp": 1765705291669,
1647
+ "user": {
1648
+ "displayName": "Anant Tripathi",
1649
+ "userId": "05588283814303116545"
1650
+ },
1651
+ "user_tz": -330
1652
+ },
1653
+ "id": "qXS_wjMdOtKM"
1654
+ },
1655
+ "outputs": [],
1656
+ "source": []
1657
+ },
1658
+ {
1659
+ "cell_type": "markdown",
1660
+ "metadata": {
1661
+ "id": "IidBPl95OtKM"
1662
+ },
1663
+ "source": [
1664
+ "![image.png](attachment:2a952d84-5287-492d-99cd-f5486c138f0e.png)"
1665
+ ]
1666
+ },
1667
+ {
1668
+ "cell_type": "markdown",
1669
+ "metadata": {
1670
+ "id": "P8QouFeUOtKM"
1671
+ },
1672
+ "source": [
1673
+ "![image.png](attachment:13c64e4a-7f76-49ae-82bc-ce339e99accc.png)"
1674
+ ]
1675
+ },
1676
+ {
1677
+ "cell_type": "markdown",
1678
+ "metadata": {
1679
+ "id": "3KDN31V2_YSr"
1680
+ },
1681
+ "source": [
1682
+ "- Streamlit on Hugging Face (link to HF space, screenshot of Streamlit app)"
1683
+ ]
1684
+ },
1685
+ {
1686
+ "cell_type": "code",
1687
+ "execution_count": 20,
1688
+ "metadata": {
1689
+ "executionInfo": {
1690
+ "elapsed": 1,
1691
+ "status": "ok",
1692
+ "timestamp": 1765705291669,
1693
+ "user": {
1694
+ "displayName": "Anant Tripathi",
1695
+ "userId": "05588283814303116545"
1696
+ },
1697
+ "user_tz": -330
1698
+ },
1699
+ "id": "NuIUdj3b_ZYV"
1700
+ },
1701
+ "outputs": [],
1702
+ "source": [
1703
+ "# TODO: Add Screenshot After Deployment\n",
1704
+ "#\n",
1705
+ "# After the workflow completes and app is deployed, add a screenshot showing:\n",
1706
+ "# 1. Live Streamlit app running on Hugging Face Spaces\n",
1707
+ "# 2. Making a sample prediction with the interface\n",
1708
+ "# 3. Prediction results displayed with confidence scores\n",
1709
+ "#\n",
1710
+ "# Also provide the Hugging Face Space URL:\n",
1711
+ "# Expected format: https://huggingface.co/spaces/<your-username>/wellness-tourism-prediction\n",
1712
+ "#\n",
1713
+ "# Screenshot should show:\n",
1714
+ "# ✅ App interface with input forms\n",
1715
+ "# ✅ Prediction button\n",
1716
+ "# ✅ Results with confidence scores\n",
1717
+ "# ✅ Recommendations"
1718
+ ]
1719
+ },
1720
+ {
1721
+ "cell_type": "markdown",
1722
+ "metadata": {
1723
+ "id": "tp06I8AnOtKM"
1724
+ },
1725
+ "source": [
1726
+ "![image.png](attachment:67adea96-d2e2-49a4-b74b-1bcb763ddf53.png)"
1727
+ ]
1728
+ },
1729
+ {
1730
+ "cell_type": "code",
1731
+ "execution_count": 20,
1732
+ "metadata": {
1733
+ "executionInfo": {
1734
+ "elapsed": 1,
1735
+ "status": "ok",
1736
+ "timestamp": 1765705291669,
1737
+ "user": {
1738
+ "displayName": "Anant Tripathi",
1739
+ "userId": "05588283814303116545"
1740
+ },
1741
+ "user_tz": -330
1742
+ },
1743
+ "id": "mK_xcFxlOtKM"
1744
+ },
1745
+ "outputs": [],
1746
+ "source": []
1747
+ },
1748
+ {
1749
+ "cell_type": "markdown",
1750
+ "metadata": {
1751
+ "id": "zWAag1sAOtKM"
1752
+ },
1753
+ "source": [
1754
+ "![image.png](attachment:2ef2ba58-64cb-4cc8-b9bb-957f4e9e5688.png)"
1755
+ ]
1756
+ },
1757
+ {
1758
+ "cell_type": "markdown",
1759
+ "metadata": {
1760
+ "id": "fN8j9-3nW8G9"
1761
+ },
1762
+ "source": [
1763
+ "<font size=6 color=\"navyblue\">Power Ahead!</font>\n",
1764
+ "___"
1765
+ ]
1766
+ }
1767
+ ],
1768
+ "metadata": {
1769
+ "colab": {
1770
+ "provenance": []
1771
+ },
1772
+ "kernel_info": {
1773
+ "name": "python310-sdkv2"
1774
+ },
1775
+ "kernelspec": {
1776
+ "display_name": "Python [conda env:anaconda3] *",
1777
+ "language": "python",
1778
+ "name": "conda-env-anaconda3-py"
1779
+ },
1780
+ "language_info": {
1781
+ "codemirror_mode": {
1782
+ "name": "ipython",
1783
+ "version": 3
1784
+ },
1785
+ "file_extension": ".py",
1786
+ "mimetype": "text/x-python",
1787
+ "name": "python",
1788
+ "nbconvert_exporter": "python",
1789
+ "pygments_lexer": "ipython3",
1790
+ "version": "3.11.4"
1791
+ },
1792
+ "microsoft": {
1793
+ "host": {
1794
+ "AzureML": {
1795
+ "notebookHasBeenCompleted": true
1796
+ }
1797
+ },
1798
+ "ms_spell_check": {
1799
+ "ms_spell_check_language": "en"
1800
+ }
1801
+ },
1802
+ "nteract": {
1803
+ "version": "nteract-front-end@1.0.0"
1804
+ }
1805
+ },
1806
+ "nbformat": 4,
1807
+ "nbformat_minor": 0
1808
+ }
Notebook_AML_and_MLOps_Project.ipynb ADDED
@@ -0,0 +1,1557 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "markdown",
5
+ "metadata": {
6
+ "id": "klg2JF-oBblG"
7
+ },
8
+ "source": [
9
+ "# Problem Statement"
10
+ ]
11
+ },
12
+ {
13
+ "cell_type": "markdown",
14
+ "metadata": {
15
+ "id": "m0CcOjZ-BblL"
16
+ },
17
+ "source": [
18
+ "## **Business Context**"
19
+ ]
20
+ },
21
+ {
22
+ "cell_type": "markdown",
23
+ "metadata": {
24
+ "id": "uyT6Koe7BblM"
25
+ },
26
+ "source": [
27
+ "\"Visit with Us,\" a leading travel company, is revolutionizing the tourism industry by leveraging data-driven strategies to optimize operations and customer engagement. While introducing a new package offering, such as the Wellness Tourism Package, the company faces challenges in targeting the right customers efficiently. The manual approach to identifying potential customers is inconsistent, time-consuming, and prone to errors, leading to missed opportunities and suboptimal campaign performance.\n",
28
+ "\n",
29
+ "To address these issues, the company aims to implement a scalable and automated system that integrates customer data, predicts potential buyers, and enhances decision-making for marketing strategies. By utilizing an MLOps pipeline, the company seeks to achieve seamless integration of data preprocessing, model development, deployment, and CI/CD practices for continuous improvement. This system will ensure efficient targeting of customers, timely updates to the predictive model, and adaptation to evolving customer behaviors, ultimately driving growth and customer satisfaction.\n"
30
+ ]
31
+ },
32
+ {
33
+ "cell_type": "markdown",
34
+ "metadata": {
35
+ "id": "zm6bNQOJBblO"
36
+ },
37
+ "source": [
38
+ "## **Objective**"
39
+ ]
40
+ },
41
+ {
42
+ "cell_type": "markdown",
43
+ "metadata": {
44
+ "id": "7PYtjk_YBblO"
45
+ },
46
+ "source": [
47
+ "As an MLOps Engineer at \"Visit with Us,\" your responsibility is to design and deploy an MLOps pipeline on GitHub to automate the end-to-end workflow for predicting customer purchases. The primary objective is to build a model that predicts whether a customer will purchase the newly introduced Wellness Tourism Package before contacting them. The pipeline will include data cleaning, preprocessing, transformation, model building, training, evaluation, and deployment, ensuring consistent performance and scalability. By leveraging GitHub Actions for CI/CD integration, the system will enable automated updates, streamline model deployment, and improve operational efficiency. This robust predictive solution will empower policymakers to make data-driven decisions, enhance marketing strategies, and effectively target potential customers, thereby driving customer acquisition and business growth."
48
+ ]
49
+ },
50
+ {
51
+ "cell_type": "markdown",
52
+ "metadata": {
53
+ "id": "z8C11AzTBblP"
54
+ },
55
+ "source": [
56
+ "## **Data Description**"
57
+ ]
58
+ },
59
+ {
60
+ "cell_type": "markdown",
61
+ "metadata": {
62
+ "id": "9DQx3pkaBblP"
63
+ },
64
+ "source": [
65
+ "The dataset contains customer and interaction data that serve as key attributes for predicting the likelihood of purchasing the Wellness Tourism Package. The detailed attributes are:\n",
66
+ "\n",
67
+ "**Customer Details**\n",
68
+ "- **CustomerID:** Unique identifier for each customer.\n",
69
+ "- **ProdTaken:** Target variable indicating whether the customer has purchased a package (0: No, 1: Yes).\n",
70
+ "- **Age:** Age of the customer.\n",
71
+ "- **TypeofContact:** The method by which the customer was contacted (Company Invited or Self Inquiry).\n",
72
+ "- **CityTier:** The city category based on development, population, and living standards (Tier 1 > Tier 2 > Tier 3).\n",
73
+ "- **Occupation:** Customer's occupation (e.g., Salaried, Freelancer).\n",
74
+ "- **Gender:** Gender of the customer (Male, Female).\n",
75
+ "- **NumberOfPersonVisiting:** Total number of people accompanying the customer on the trip.\n",
76
+ "- **PreferredPropertyStar:** Preferred hotel rating by the customer.\n",
77
+ "- **MaritalStatus:** Marital status of the customer (Single, Married, Divorced).\n",
78
+ "- **NumberOfTrips:** Average number of trips the customer takes annually.\n",
79
+ "- **Passport:** Whether the customer holds a valid passport (0: No, 1: Yes).\n",
80
+ "- **OwnCar:** Whether the customer owns a car (0: No, 1: Yes).\n",
81
+ "- **NumberOfChildrenVisiting:** Number of children below age 5 accompanying the customer.\n",
82
+ "- **Designation:** Customer's designation in their current organization.\n",
83
+ "- **MonthlyIncome:** Gross monthly income of the customer.\n",
84
+ "\n",
85
+ "**Customer Interaction Data**\n",
86
+ "- **PitchSatisfactionScore:** Score indicating the customer's satisfaction with the sales pitch.\n",
87
+ "- **ProductPitched:** The type of product pitched to the customer.\n",
88
+ "- **NumberOfFollowups:** Total number of follow-ups by the salesperson after the sales pitch.-\n",
89
+ "- **DurationOfPitch:** Duration of the sales pitch delivered to the customer.\n"
90
+ ]
91
+ },
92
+ {
93
+ "cell_type": "markdown",
94
+ "metadata": {
95
+ "id": "0LbSu_p2jYfe"
96
+ },
97
+ "source": [
98
+ "# Model Building"
99
+ ]
100
+ },
101
+ {
102
+ "cell_type": "code",
103
+ "execution_count": 1,
104
+ "metadata": {
105
+ "colab": {
106
+ "base_uri": "https://colab.research.google.com/"
107
+ },
108
+ "id": "giodc4KknHID"
109
+ },
110
+ "outputs": [
111
+ {
112
+ "name": "stdout",
113
+ "output_type": "stream",
114
+ "text": [
115
+ "/content\n"
116
+ ]
117
+ }
118
+ ],
119
+ "source": [
120
+ "# Create a master folder to keep all files created when executing the below code cells\n",
121
+ "import os\n",
122
+ "print(os.getcwd())\n",
123
+ "os.makedirs(\"tourism_project\", exist_ok=True)"
124
+ ]
125
+ },
126
+ {
127
+ "cell_type": "code",
128
+ "execution_count": 2,
129
+ "metadata": {
130
+ "id": "SUKPoy0EA4jj"
131
+ },
132
+ "outputs": [],
133
+ "source": [
134
+ "# Create a folder for storing the model building files\n",
135
+ "os.makedirs(\"tourism_project/model_building\", exist_ok=True)"
136
+ ]
137
+ },
138
+ {
139
+ "cell_type": "markdown",
140
+ "metadata": {
141
+ "id": "9DtS3gNDjBbR"
142
+ },
143
+ "source": [
144
+ "## Data Registration"
145
+ ]
146
+ },
147
+ {
148
+ "cell_type": "code",
149
+ "execution_count": 3,
150
+ "metadata": {
151
+ "id": "ZagOeVxJOtJ9"
152
+ },
153
+ "outputs": [],
154
+ "source": [
155
+ "os.makedirs(\"tourism_project/data\", exist_ok=True)"
156
+ ]
157
+ },
158
+ {
159
+ "cell_type": "markdown",
160
+ "metadata": {
161
+ "id": "WxXiD9ZXxodF"
162
+ },
163
+ "source": [
164
+ "Once the **data** folder created after executing the above cell, please upload the **tourism.csv** in to the folder"
165
+ ]
166
+ },
167
+ {
168
+ "cell_type": "markdown",
169
+ "metadata": {
170
+ "id": "hh2TjRG5WJ4Z"
171
+ },
172
+ "source": [
173
+ "## Data Preparation"
174
+ ]
175
+ },
176
+ {
177
+ "cell_type": "code",
178
+ "execution_count": 4,
179
+ "metadata": {
180
+ "colab": {
181
+ "base_uri": "https://colab.research.google.com/"
182
+ },
183
+ "id": "EHVRGAeoOtJ-"
184
+ },
185
+ "outputs": [
186
+ {
187
+ "name": "stdout",
188
+ "output_type": "stream",
189
+ "text": [
190
+ "Writing tourism_project/model_building/data_register.py\n"
191
+ ]
192
+ }
193
+ ],
194
+ "source": [
195
+ "%%writefile tourism_project/model_building/data_register.py\n",
196
+ "from huggingface_hub.utils import RepositoryNotFoundError, HfHubHTTPError\n",
197
+ "from huggingface_hub import HfApi, create_repo\n",
198
+ "import os\n",
199
+ "\n",
200
+ "\n",
201
+ "repo_id = \"ananttripathiak/tourism-dataset\"\n",
202
+ "repo_type = \"dataset\"\n",
203
+ "\n",
204
+ "# Initialize API client\n",
205
+ "api = HfApi(token=os.getenv(\"HF_TOKEN\"))\n",
206
+ "\n",
207
+ "# Step 1: Check if the space exists\n",
208
+ "try:\n",
209
+ " api.repo_info(repo_id=repo_id, repo_type=repo_type)\n",
210
+ " print(f\"Space '{repo_id}' already exists. Using it.\")\n",
211
+ "except RepositoryNotFoundError:\n",
212
+ " print(f\"Space '{repo_id}' not found. Creating new space...\")\n",
213
+ " create_repo(repo_id=repo_id, repo_type=repo_type, private=False)\n",
214
+ " print(f\"Space '{repo_id}' created.\")\n",
215
+ "\n",
216
+ "api.upload_folder(\n",
217
+ " folder_path=\"tourism_project/data\",\n",
218
+ " repo_id=repo_id,\n",
219
+ " repo_type=repo_type,\n",
220
+ ")"
221
+ ]
222
+ },
223
+ {
224
+ "cell_type": "markdown",
225
+ "metadata": {
226
+ "id": "eZZKnLkLjeM4"
227
+ },
228
+ "source": [
229
+ "## Model Training and Registration with Experimentation Tracking"
230
+ ]
231
+ },
232
+ {
233
+ "cell_type": "code",
234
+ "execution_count": 5,
235
+ "metadata": {
236
+ "id": "LFmrcXT_OtJ-"
237
+ },
238
+ "outputs": [],
239
+ "source": [
240
+ "# !pip install mlflow scikit-learn huggingface_hub\n",
241
+ "# import sys\n",
242
+ "# !{sys.executable} -m pip install mlflow"
243
+ ]
244
+ },
245
+ {
246
+ "cell_type": "code",
247
+ "execution_count": 6,
248
+ "metadata": {
249
+ "colab": {
250
+ "base_uri": "https://colab.research.google.com/"
251
+ },
252
+ "id": "DatpH_YdOtJ-"
253
+ },
254
+ "outputs": [
255
+ {
256
+ "name": "stdout",
257
+ "output_type": "stream",
258
+ "text": [
259
+ "Writing tourism_project/model_building/train.py\n"
260
+ ]
261
+ }
262
+ ],
263
+ "source": [
264
+ "%%writefile tourism_project/model_building/train.py\n",
265
+ "\n",
266
+ "# for data manipulation\n",
267
+ "import pandas as pd\n",
268
+ "import numpy as np\n",
269
+ "# for data preprocessing and pipeline creation\n",
270
+ "from sklearn.preprocessing import StandardScaler\n",
271
+ "from sklearn.compose import make_column_transformer\n",
272
+ "from sklearn.pipeline import make_pipeline\n",
273
+ "# for model training, tuning, and evaluation\n",
274
+ "import xgboost as xgb\n",
275
+ "from sklearn.model_selection import GridSearchCV\n",
276
+ "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, roc_auc_score, classification_report, confusion_matrix\n",
277
+ "# for model serialization\n",
278
+ "import joblib\n",
279
+ "# for creating a folder\n",
280
+ "import os\n",
281
+ "# for hugging face space authentication to upload files\n",
282
+ "from huggingface_hub import login, HfApi, create_repo\n",
283
+ "from huggingface_hub.utils import RepositoryNotFoundError, HfHubHTTPError\n",
284
+ "import mlflow\n",
285
+ "\n",
286
+ "# Set up MLflow tracking\n",
287
+ "mlflow.set_tracking_uri(\"http://localhost:5000\")\n",
288
+ "mlflow.set_experiment(\"tourism-package-prediction\")\n",
289
+ "\n",
290
+ "api = HfApi()\n",
291
+ "\n",
292
+ "# Load the preprocessed data from Hugging Face\n",
293
+ "Xtrain_path = \"hf://datasets/ananttripathiak/tourism-dataset/Xtrain.csv\"\n",
294
+ "Xtest_path = \"hf://datasets/ananttripathiak/tourism-dataset/Xtest.csv\"\n",
295
+ "ytrain_path = \"hf://datasets/ananttripathiak/tourism-dataset/ytrain.csv\"\n",
296
+ "ytest_path = \"hf://datasets/ananttripathiak/tourism-dataset/ytest.csv\"\n",
297
+ "\n",
298
+ "print(\"Loading preprocessed data...\")\n",
299
+ "Xtrain = pd.read_csv(Xtrain_path)\n",
300
+ "Xtest = pd.read_csv(Xtest_path)\n",
301
+ "ytrain = pd.read_csv(ytrain_path).values.ravel()\n",
302
+ "ytest = pd.read_csv(ytest_path).values.ravel()\n",
303
+ "\n",
304
+ "print(f\"Training set shape: {Xtrain.shape}\")\n",
305
+ "print(f\"Test set shape: {Xtest.shape}\")\n",
306
+ "\n",
307
+ "# Identify numeric features (all features after encoding)\n",
308
+ "numeric_features = Xtrain.columns.tolist()\n",
309
+ "\n",
310
+ "# Preprocessor - StandardScaler for all numeric features\n",
311
+ "preprocessor = make_column_transformer(\n",
312
+ " (StandardScaler(), numeric_features)\n",
313
+ ")\n",
314
+ "\n",
315
+ "# Define base XGBoost Classifier\n",
316
+ "xgb_model = xgb.XGBClassifier(\n",
317
+ " random_state=42,\n",
318
+ " n_jobs=-1,\n",
319
+ " eval_metric='logloss',\n",
320
+ " use_label_encoder=False\n",
321
+ ")\n",
322
+ "\n",
323
+ "# Hyperparameter grid for classification\n",
324
+ "param_grid = {\n",
325
+ " 'xgbclassifier__n_estimators': [100, 200, 300],\n",
326
+ " 'xgbclassifier__max_depth': [3, 5, 7],\n",
327
+ " 'xgbclassifier__learning_rate': [0.01, 0.05, 0.1],\n",
328
+ " 'xgbclassifier__subsample': [0.7, 0.8, 1.0],\n",
329
+ " 'xgbclassifier__colsample_bytree': [0.7, 0.8, 1.0],\n",
330
+ " 'xgbclassifier__scale_pos_weight': [1, 2, 3] # Handle class imbalance\n",
331
+ "}\n",
332
+ "\n",
333
+ "# Pipeline\n",
334
+ "model_pipeline = make_pipeline(preprocessor, xgb_model)\n",
335
+ "\n",
336
+ "print(\"\\nStarting MLflow experiment...\")\n",
337
+ "with mlflow.start_run():\n",
338
+ " print(\"Performing Grid Search with Cross-Validation...\")\n",
339
+ " # Grid Search\n",
340
+ " grid_search = GridSearchCV(\n",
341
+ " model_pipeline,\n",
342
+ " param_grid,\n",
343
+ " cv=3,\n",
344
+ " n_jobs=-1,\n",
345
+ " scoring='roc_auc',\n",
346
+ " verbose=1\n",
347
+ " )\n",
348
+ " grid_search.fit(Xtrain, ytrain)\n",
349
+ "\n",
350
+ " # Log parameter sets\n",
351
+ " results = grid_search.cv_results_\n",
352
+ " print(f\"\\nEvaluated {len(results['params'])} parameter combinations\")\n",
353
+ "\n",
354
+ " for i in range(len(results['params'])):\n",
355
+ " param_set = results['params'][i]\n",
356
+ " mean_score = results['mean_test_score'][i]\n",
357
+ "\n",
358
+ " with mlflow.start_run(nested=True):\n",
359
+ " mlflow.log_params(param_set)\n",
360
+ " mlflow.log_metric(\"mean_roc_auc\", mean_score)\n",
361
+ "\n",
362
+ " # Best model\n",
363
+ " print(f\"\\nBest parameters: {grid_search.best_params_}\")\n",
364
+ " mlflow.log_params(grid_search.best_params_)\n",
365
+ " best_model = grid_search.best_estimator_\n",
366
+ "\n",
367
+ " # Predictions\n",
368
+ " print(\"\\nMaking predictions...\")\n",
369
+ " y_pred_train = best_model.predict(Xtrain)\n",
370
+ " y_pred_test = best_model.predict(Xtest)\n",
371
+ "\n",
372
+ " # Probability predictions for ROC-AUC\n",
373
+ " y_pred_train_proba = best_model.predict_proba(Xtrain)[:, 1]\n",
374
+ " y_pred_test_proba = best_model.predict_proba(Xtest)[:, 1]\n",
375
+ "\n",
376
+ " # Calculate metrics\n",
377
+ " print(\"\\nCalculating metrics...\")\n",
378
+ " train_accuracy = accuracy_score(ytrain, y_pred_train)\n",
379
+ " test_accuracy = accuracy_score(ytest, y_pred_test)\n",
380
+ "\n",
381
+ " train_precision = precision_score(ytrain, y_pred_train, zero_division=0)\n",
382
+ " test_precision = precision_score(ytest, y_pred_test, zero_division=0)\n",
383
+ "\n",
384
+ " train_recall = recall_score(ytrain, y_pred_train, zero_division=0)\n",
385
+ " test_recall = recall_score(ytest, y_pred_test, zero_division=0)\n",
386
+ "\n",
387
+ " train_f1 = f1_score(ytrain, y_pred_train, zero_division=0)\n",
388
+ " test_f1 = f1_score(ytest, y_pred_test, zero_division=0)\n",
389
+ "\n",
390
+ " train_roc_auc = roc_auc_score(ytrain, y_pred_train_proba)\n",
391
+ " test_roc_auc = roc_auc_score(ytest, y_pred_test_proba)\n",
392
+ "\n",
393
+ " # Log metrics\n",
394
+ " mlflow.log_metrics({\n",
395
+ " \"train_accuracy\": train_accuracy,\n",
396
+ " \"test_accuracy\": test_accuracy,\n",
397
+ " \"train_precision\": train_precision,\n",
398
+ " \"test_precision\": test_precision,\n",
399
+ " \"train_recall\": train_recall,\n",
400
+ " \"test_recall\": test_recall,\n",
401
+ " \"train_f1_score\": train_f1,\n",
402
+ " \"test_f1_score\": test_f1,\n",
403
+ " \"train_roc_auc\": train_roc_auc,\n",
404
+ " \"test_roc_auc\": test_roc_auc\n",
405
+ " })\n",
406
+ "\n",
407
+ " # Print results\n",
408
+ " print(\"\\n\" + \"=\"*50)\n",
409
+ " print(\"MODEL PERFORMANCE METRICS\")\n",
410
+ " print(\"=\"*50)\n",
411
+ " print(f\"Train Accuracy: {train_accuracy:.4f} | Test Accuracy: {test_accuracy:.4f}\")\n",
412
+ " print(f\"Train Precision: {train_precision:.4f} | Test Precision: {test_precision:.4f}\")\n",
413
+ " print(f\"Train Recall: {train_recall:.4f} | Test Recall: {test_recall:.4f}\")\n",
414
+ " print(f\"Train F1-Score: {train_f1:.4f} | Test F1-Score: {test_f1:.4f}\")\n",
415
+ " print(f\"Train ROC-AUC: {train_roc_auc:.4f} | Test ROC-AUC: {test_roc_auc:.4f}\")\n",
416
+ " print(\"=\"*50)\n",
417
+ "\n",
418
+ " print(\"\\nTest Set Classification Report:\")\n",
419
+ " print(classification_report(ytest, y_pred_test, target_names=['No Purchase', 'Purchase']))\n",
420
+ "\n",
421
+ " print(\"\\nTest Set Confusion Matrix:\")\n",
422
+ " print(confusion_matrix(ytest, y_pred_test))\n",
423
+ "\n",
424
+ " # Save the model locally\n",
425
+ " model_path = \"best_tourism_model_v1.joblib\"\n",
426
+ " joblib.dump(best_model, model_path)\n",
427
+ " print(f\"\\nModel saved locally as: {model_path}\")\n",
428
+ "\n",
429
+ " # Log the model artifact\n",
430
+ " mlflow.log_artifact(model_path, artifact_path=\"model\")\n",
431
+ " print(f\"Model logged to MLflow\")\n",
432
+ "\n",
433
+ " # Upload to Hugging Face\n",
434
+ " repo_id = \"ananttripathiak/tourism-prediction-model\"\n",
435
+ " repo_type = \"model\"\n",
436
+ "\n",
437
+ " # Step 1: Check if the repository exists\n",
438
+ " try:\n",
439
+ " api.repo_info(repo_id=repo_id, repo_type=repo_type)\n",
440
+ " print(f\"\\nRepository '{repo_id}' already exists. Using it.\")\n",
441
+ " except RepositoryNotFoundError:\n",
442
+ " print(f\"\\nRepository '{repo_id}' not found. Creating new repository...\")\n",
443
+ " create_repo(repo_id=repo_id, repo_type=repo_type, private=False)\n",
444
+ " print(f\"Repository '{repo_id}' created.\")\n",
445
+ "\n",
446
+ " # Upload model to Hugging Face\n",
447
+ " api.upload_file(\n",
448
+ " path_or_fileobj=\"best_tourism_model_v1.joblib\",\n",
449
+ " path_in_repo=\"best_tourism_model_v1.joblib\",\n",
450
+ " repo_id=repo_id,\n",
451
+ " repo_type=repo_type,\n",
452
+ " )\n",
453
+ " print(f\"Model uploaded to Hugging Face: {repo_id}\")\n",
454
+ "\n",
455
+ "print(\"\\n\" + \"=\"*50)\n",
456
+ "print(\"MODEL TRAINING COMPLETED SUCCESSFULLY!\")\n",
457
+ "print(\"=\"*50)"
458
+ ]
459
+ },
460
+ {
461
+ "cell_type": "code",
462
+ "execution_count": 7,
463
+ "metadata": {
464
+ "colab": {
465
+ "base_uri": "https://colab.research.google.com/"
466
+ },
467
+ "id": "Vtjn_63uOtJ-"
468
+ },
469
+ "outputs": [
470
+ {
471
+ "name": "stdout",
472
+ "output_type": "stream",
473
+ "text": [
474
+ "Writing tourism_project/model_building/prep.py\n"
475
+ ]
476
+ }
477
+ ],
478
+ "source": [
479
+ "%%writefile tourism_project/model_building/prep.py\n",
480
+ "# for data manipulation\n",
481
+ "import pandas as pd\n",
482
+ "import numpy as np\n",
483
+ "# for data preprocessing and pipeline creation\n",
484
+ "from sklearn.model_selection import train_test_split\n",
485
+ "# for converting text data in to numerical representation\n",
486
+ "from sklearn.preprocessing import LabelEncoder\n",
487
+ "# for hugging face space authentication to upload files\n",
488
+ "from huggingface_hub import login, HfApi\n",
489
+ "import os\n",
490
+ "\n",
491
+ "# Define constants for the dataset and output paths\n",
492
+ "api = HfApi(token=os.getenv(\"HF_TOKEN\"))\n",
493
+ "DATASET_PATH = \"hf://datasets/ananttripathiak/tourism-dataset/tourism.csv\"\n",
494
+ "df = pd.read_csv(DATASET_PATH)\n",
495
+ "print(\"Dataset loaded successfully.\")\n",
496
+ "print(f\"Dataset shape: {df.shape}\")\n",
497
+ "\n",
498
+ "# Drop the unnamed index column if it exists\n",
499
+ "if 'Unnamed: 0' in df.columns or df.columns[0] == '':\n",
500
+ " df = df.iloc[:, 1:]\n",
501
+ "\n",
502
+ "# Drop CustomerID as it's a unique identifier (not useful for modeling)\n",
503
+ "if 'CustomerID' in df.columns:\n",
504
+ " df.drop(columns=['CustomerID'], inplace=True)\n",
505
+ "\n",
506
+ "# Handle missing values\n",
507
+ "print(\"\\nHandling missing values...\")\n",
508
+ "# For numerical columns, fill with median\n",
509
+ "numerical_cols = df.select_dtypes(include=[np.number]).columns\n",
510
+ "for col in numerical_cols:\n",
511
+ " if df[col].isnull().sum() > 0:\n",
512
+ " df[col].fillna(df[col].median(), inplace=True)\n",
513
+ "\n",
514
+ "# For categorical columns, fill with mode\n",
515
+ "categorical_cols = df.select_dtypes(include=['object']).columns\n",
516
+ "for col in categorical_cols:\n",
517
+ " if df[col].isnull().sum() > 0:\n",
518
+ " df[col].fillna(df[col].mode()[0], inplace=True)\n",
519
+ "\n",
520
+ "# Handle specific data quality issues (e.g., \"Fe Male\" should be \"Female\")\n",
521
+ "if 'Gender' in df.columns:\n",
522
+ " df['Gender'] = df['Gender'].str.strip().replace({'Fe Male': 'Female', 'Fe male': 'Female'})\n",
523
+ "\n",
524
+ "# Encode categorical columns\n",
525
+ "print(\"\\nEncoding categorical variables...\")\n",
526
+ "label_encoder = LabelEncoder()\n",
527
+ "\n",
528
+ "# List of categorical columns to encode\n",
529
+ "categorical_features = ['TypeofContact', 'Occupation', 'Gender', 'ProductPitched',\n",
530
+ " 'MaritalStatus', 'Designation']\n",
531
+ "\n",
532
+ "for col in categorical_features:\n",
533
+ " if col in df.columns:\n",
534
+ " df[col] = label_encoder.fit_transform(df[col].astype(str))\n",
535
+ "\n",
536
+ "# Define target variable\n",
537
+ "target_col = 'ProdTaken'\n",
538
+ "\n",
539
+ "# Split into X (features) and y (target)\n",
540
+ "X = df.drop(columns=[target_col])\n",
541
+ "y = df[target_col]\n",
542
+ "\n",
543
+ "print(f\"\\nFeatures shape: {X.shape}\")\n",
544
+ "print(f\"Target shape: {y.shape}\")\n",
545
+ "print(f\"Target distribution:\\n{y.value_counts()}\")\n",
546
+ "\n",
547
+ "# Perform train-test split\n",
548
+ "Xtrain, Xtest, ytrain, ytest = train_test_split(\n",
549
+ " X, y, test_size=0.2, random_state=42, stratify=y\n",
550
+ ")\n",
551
+ "\n",
552
+ "print(f\"\\nTrain set size: {Xtrain.shape[0]}\")\n",
553
+ "print(f\"Test set size: {Xtest.shape[0]}\")\n",
554
+ "\n",
555
+ "# Save the datasets\n",
556
+ "Xtrain.to_csv(\"Xtrain.csv\", index=False)\n",
557
+ "Xtest.to_csv(\"Xtest.csv\", index=False)\n",
558
+ "ytrain.to_csv(\"ytrain.csv\", index=False)\n",
559
+ "ytest.to_csv(\"ytest.csv\", index=False)\n",
560
+ "\n",
561
+ "print(\"\\nDatasets saved locally.\")\n",
562
+ "\n",
563
+ "# Upload to Hugging Face\n",
564
+ "files = [\"Xtrain.csv\", \"Xtest.csv\", \"ytrain.csv\", \"ytest.csv\"]\n",
565
+ "\n",
566
+ "for file_path in files:\n",
567
+ " api.upload_file(\n",
568
+ " path_or_fileobj=file_path,\n",
569
+ " path_in_repo=file_path.split(\"/\")[-1],\n",
570
+ " repo_id=\"ananttripathiak/tourism-dataset\",\n",
571
+ " repo_type=\"dataset\",\n",
572
+ " )\n",
573
+ " print(f\"Uploaded {file_path} to Hugging Face\")\n",
574
+ "\n",
575
+ "print(\"\\nData preparation completed successfully!\")"
576
+ ]
577
+ },
578
+ {
579
+ "cell_type": "markdown",
580
+ "metadata": {
581
+ "id": "0McYCZzkji5I"
582
+ },
583
+ "source": [
584
+ "# Deployment"
585
+ ]
586
+ },
587
+ {
588
+ "cell_type": "markdown",
589
+ "metadata": {
590
+ "id": "9QrY2v77vbEZ"
591
+ },
592
+ "source": [
593
+ "## Dockerfile"
594
+ ]
595
+ },
596
+ {
597
+ "cell_type": "code",
598
+ "execution_count": 8,
599
+ "metadata": {
600
+ "id": "0-AMAI72CR-T"
601
+ },
602
+ "outputs": [],
603
+ "source": [
604
+ "os.makedirs(\"tourism_project/deployment\", exist_ok=True)"
605
+ ]
606
+ },
607
+ {
608
+ "cell_type": "code",
609
+ "execution_count": 9,
610
+ "metadata": {
611
+ "colab": {
612
+ "base_uri": "https://colab.research.google.com/"
613
+ },
614
+ "id": "ZTicTDnPCVZr"
615
+ },
616
+ "outputs": [
617
+ {
618
+ "name": "stdout",
619
+ "output_type": "stream",
620
+ "text": [
621
+ "Writing tourism_project/deployment/Dockerfile\n"
622
+ ]
623
+ }
624
+ ],
625
+ "source": [
626
+ "%%writefile tourism_project/deployment/Dockerfile\n",
627
+ "# Use a minimal base image with Python 3.9 installed\n",
628
+ "FROM python:3.9\n",
629
+ "\n",
630
+ "# Set the working directory inside the container to /app\n",
631
+ "WORKDIR /app\n",
632
+ "\n",
633
+ "# Copy all files from the current directory on the host to the container's /app directory\n",
634
+ "COPY . .\n",
635
+ "\n",
636
+ "# Install Python dependencies listed in requirements.txt\n",
637
+ "RUN pip3 install -r requirements.txt\n",
638
+ "\n",
639
+ "RUN useradd -m -u 1000 user\n",
640
+ "USER user\n",
641
+ "ENV HOME=/home/user \\\n",
642
+ "\tPATH=/home/user/.local/bin:$PATH\n",
643
+ "\n",
644
+ "WORKDIR $HOME/app\n",
645
+ "\n",
646
+ "COPY --chown=user . $HOME/app\n",
647
+ "\n",
648
+ "# Define the command to run the Streamlit app on port \"8501\" and make it accessible externally\n",
649
+ "CMD [\"streamlit\", \"run\", \"app.py\", \"--server.port=8501\", \"--server.address=0.0.0.0\", \"--server.enableXsrfProtection=false\"]"
650
+ ]
651
+ },
652
+ {
653
+ "cell_type": "markdown",
654
+ "metadata": {
655
+ "id": "LCvrklrBwNvJ"
656
+ },
657
+ "source": [
658
+ "## Streamlit App"
659
+ ]
660
+ },
661
+ {
662
+ "cell_type": "markdown",
663
+ "metadata": {
664
+ "id": "fXWe6ObRjP6-"
665
+ },
666
+ "source": [
667
+ "Please ensure that the web app script is named `app.py`."
668
+ ]
669
+ },
670
+ {
671
+ "cell_type": "code",
672
+ "execution_count": 10,
673
+ "metadata": {
674
+ "colab": {
675
+ "base_uri": "https://colab.research.google.com/"
676
+ },
677
+ "id": "WBG-jxM89jdp"
678
+ },
679
+ "outputs": [
680
+ {
681
+ "name": "stdout",
682
+ "output_type": "stream",
683
+ "text": [
684
+ "Writing tourism_project/deployment/app.py\n"
685
+ ]
686
+ }
687
+ ],
688
+ "source": [
689
+ "%%writefile tourism_project/deployment/app.py\n",
690
+ "import streamlit as st\n",
691
+ "import pandas as pd\n",
692
+ "import numpy as np\n",
693
+ "from huggingface_hub import hf_hub_download\n",
694
+ "import joblib\n",
695
+ "\n",
696
+ "# Page configuration\n",
697
+ "st.set_page_config(\n",
698
+ " page_title=\"Tourism Package Prediction\",\n",
699
+ " page_icon=\"✈️\",\n",
700
+ " layout=\"wide\"\n",
701
+ ")\n",
702
+ "\n",
703
+ "# Download and load the trained model\n",
704
+ "@st.cache_resource\n",
705
+ "def load_model():\n",
706
+ " try:\n",
707
+ " model_path = hf_hub_download(\n",
708
+ " repo_id=\"ananttripathiak/tourism-prediction-model\",\n",
709
+ " filename=\"best_tourism_model_v1.joblib\"\n",
710
+ " )\n",
711
+ " model = joblib.load(model_path)\n",
712
+ " return model\n",
713
+ " except Exception as e:\n",
714
+ " st.error(f\"Error loading model: {e}\")\n",
715
+ " return None\n",
716
+ "\n",
717
+ "model = load_model()\n",
718
+ "\n",
719
+ "# Title and Description\n",
720
+ "st.title(\"✈️ Wellness Tourism Package Prediction\")\n",
721
+ "st.markdown(\"\"\"\n",
722
+ "<style>\n",
723
+ " .main-header {\n",
724
+ " font-size: 20px;\n",
725
+ " color: #1f77b4;\n",
726
+ " margin-bottom: 20px;\n",
727
+ " }\n",
728
+ " .prediction-box {\n",
729
+ " padding: 20px;\n",
730
+ " border-radius: 10px;\n",
731
+ " margin: 20px 0;\n",
732
+ " }\n",
733
+ " .success-box {\n",
734
+ " background-color: #d4edda;\n",
735
+ " border: 1px solid #c3e6cb;\n",
736
+ " }\n",
737
+ " .warning-box {\n",
738
+ " background-color: #fff3cd;\n",
739
+ " border: 1px solid #ffeeba;\n",
740
+ " }\n",
741
+ "</style>\n",
742
+ "\"\"\", unsafe_allow_html=True)\n",
743
+ "\n",
744
+ "st.markdown(\"\"\"\n",
745
+ "This application predicts the likelihood of a customer purchasing the **Wellness Tourism Package**\n",
746
+ "based on their profile and interaction data. Enter customer details below to get a prediction.\n",
747
+ "\"\"\")\n",
748
+ "\n",
749
+ "# Create two columns for input\n",
750
+ "col1, col2 = st.columns(2)\n",
751
+ "\n",
752
+ "with col1:\n",
753
+ " st.subheader(\"📋 Customer Demographics\")\n",
754
+ "\n",
755
+ " age = st.number_input(\"Age\", min_value=18, max_value=100, value=35, step=1)\n",
756
+ "\n",
757
+ " type_of_contact = st.selectbox(\n",
758
+ " \"Type of Contact\",\n",
759
+ " options=[\"Self Enquiry\", \"Company Invited\"]\n",
760
+ " )\n",
761
+ "\n",
762
+ " city_tier = st.selectbox(\n",
763
+ " \"City Tier\",\n",
764
+ " options=[1, 2, 3],\n",
765
+ " help=\"Tier 1: Metro cities, Tier 2: Mid-sized cities, Tier 3: Smaller cities\"\n",
766
+ " )\n",
767
+ "\n",
768
+ " occupation = st.selectbox(\n",
769
+ " \"Occupation\",\n",
770
+ " options=[\"Salaried\", \"Small Business\", \"Free Lancer\", \"Large Business\"]\n",
771
+ " )\n",
772
+ "\n",
773
+ " gender = st.selectbox(\"Gender\", options=[\"Male\", \"Female\"])\n",
774
+ "\n",
775
+ " marital_status = st.selectbox(\n",
776
+ " \"Marital Status\",\n",
777
+ " options=[\"Single\", \"Married\", \"Divorced\", \"Unmarried\"]\n",
778
+ " )\n",
779
+ "\n",
780
+ " designation = st.selectbox(\n",
781
+ " \"Designation\",\n",
782
+ " options=[\"Executive\", \"Manager\", \"Senior Manager\", \"AVP\", \"VP\"]\n",
783
+ " )\n",
784
+ "\n",
785
+ " monthly_income = st.number_input(\n",
786
+ " \"Monthly Income (₹)\",\n",
787
+ " min_value=0.0,\n",
788
+ " max_value=200000.0,\n",
789
+ " value=25000.0,\n",
790
+ " step=1000.0\n",
791
+ " )\n",
792
+ "\n",
793
+ "with col2:\n",
794
+ " st.subheader(\"🎯 Customer Interaction & Preferences\")\n",
795
+ "\n",
796
+ " duration_of_pitch = st.number_input(\n",
797
+ " \"Duration of Pitch (minutes)\",\n",
798
+ " min_value=0.0,\n",
799
+ " max_value=60.0,\n",
800
+ " value=15.0,\n",
801
+ " step=0.5\n",
802
+ " )\n",
803
+ "\n",
804
+ " number_of_persons_visiting = st.number_input(\n",
805
+ " \"Number of Persons Visiting\",\n",
806
+ " min_value=1,\n",
807
+ " max_value=10,\n",
808
+ " value=2,\n",
809
+ " step=1\n",
810
+ " )\n",
811
+ "\n",
812
+ " number_of_followups = st.number_input(\n",
813
+ " \"Number of Follow-ups\",\n",
814
+ " min_value=0.0,\n",
815
+ " max_value=10.0,\n",
816
+ " value=3.0,\n",
817
+ " step=1.0\n",
818
+ " )\n",
819
+ "\n",
820
+ " product_pitched = st.selectbox(\n",
821
+ " \"Product Pitched\",\n",
822
+ " options=[\"Basic\", \"Standard\", \"Deluxe\", \"Super Deluxe\", \"King\"]\n",
823
+ " )\n",
824
+ "\n",
825
+ " preferred_property_star = st.selectbox(\n",
826
+ " \"Preferred Property Star Rating\",\n",
827
+ " options=[3.0, 4.0, 5.0]\n",
828
+ " )\n",
829
+ "\n",
830
+ " number_of_trips = st.number_input(\n",
831
+ " \"Number of Trips (per year)\",\n",
832
+ " min_value=0.0,\n",
833
+ " max_value=20.0,\n",
834
+ " value=3.0,\n",
835
+ " step=1.0\n",
836
+ " )\n",
837
+ "\n",
838
+ " passport = st.selectbox(\"Has Passport?\", options=[\"Yes\", \"No\"])\n",
839
+ "\n",
840
+ " pitch_satisfaction_score = st.slider(\n",
841
+ " \"Pitch Satisfaction Score\",\n",
842
+ " min_value=1,\n",
843
+ " max_value=5,\n",
844
+ " value=3,\n",
845
+ " step=1\n",
846
+ " )\n",
847
+ "\n",
848
+ " own_car = st.selectbox(\"Owns Car?\", options=[\"Yes\", \"No\"])\n",
849
+ "\n",
850
+ " number_of_children_visiting = st.number_input(\n",
851
+ " \"Number of Children Visiting\",\n",
852
+ " min_value=0.0,\n",
853
+ " max_value=5.0,\n",
854
+ " value=0.0,\n",
855
+ " step=1.0\n",
856
+ " )\n",
857
+ "\n",
858
+ "# Encoding mapping (based on LabelEncoder used during training)\n",
859
+ "# Note: These mappings should match the exact encoding used during training\n",
860
+ "type_of_contact_map = {\"Company Invited\": 0, \"Self Enquiry\": 1}\n",
861
+ "occupation_map = {\"Free Lancer\": 0, \"Large Business\": 1, \"Salaried\": 2, \"Small Business\": 3}\n",
862
+ "gender_map = {\"Female\": 0, \"Male\": 1}\n",
863
+ "product_pitched_map = {\"Basic\": 0, \"Deluxe\": 1, \"King\": 2, \"Standard\": 3, \"Super Deluxe\": 4}\n",
864
+ "marital_status_map = {\"Divorced\": 0, \"Married\": 1, \"Single\": 2, \"Unmarried\": 3}\n",
865
+ "designation_map = {\"AVP\": 0, \"Executive\": 1, \"Manager\": 2, \"Senior Manager\": 3, \"VP\": 4}\n",
866
+ "\n",
867
+ "# Convert Yes/No to 0/1\n",
868
+ "passport_val = 1 if passport == \"Yes\" else 0\n",
869
+ "own_car_val = 1 if own_car == \"Yes\" else 0\n",
870
+ "\n",
871
+ "# Prepare input data\n",
872
+ "input_data = pd.DataFrame([{\n",
873
+ " 'Age': age,\n",
874
+ " 'TypeofContact': type_of_contact_map[type_of_contact],\n",
875
+ " 'CityTier': city_tier,\n",
876
+ " 'DurationOfPitch': duration_of_pitch,\n",
877
+ " 'Occupation': occupation_map[occupation],\n",
878
+ " 'Gender': gender_map[gender],\n",
879
+ " 'NumberOfPersonVisiting': number_of_persons_visiting,\n",
880
+ " 'NumberOfFollowups': number_of_followups,\n",
881
+ " 'ProductPitched': product_pitched_map[product_pitched],\n",
882
+ " 'PreferredPropertyStar': preferred_property_star,\n",
883
+ " 'MaritalStatus': marital_status_map[marital_status],\n",
884
+ " 'NumberOfTrips': number_of_trips,\n",
885
+ " 'Passport': passport_val,\n",
886
+ " 'PitchSatisfactionScore': pitch_satisfaction_score,\n",
887
+ " 'OwnCar': own_car_val,\n",
888
+ " 'NumberOfChildrenVisiting': number_of_children_visiting,\n",
889
+ " 'Designation': designation_map[designation],\n",
890
+ " 'MonthlyIncome': monthly_income\n",
891
+ "}])\n",
892
+ "\n",
893
+ "# Predict button\n",
894
+ "st.markdown(\"---\")\n",
895
+ "if st.button(\"🔮 Predict Purchase Likelihood\", use_container_width=True):\n",
896
+ " if model is not None:\n",
897
+ " try:\n",
898
+ " # Make prediction\n",
899
+ " prediction = model.predict(input_data)[0]\n",
900
+ " prediction_proba = model.predict_proba(input_data)[0]\n",
901
+ "\n",
902
+ " # Display results\n",
903
+ " st.markdown(\"### 📊 Prediction Results\")\n",
904
+ "\n",
905
+ " if prediction == 1:\n",
906
+ " st.markdown(\n",
907
+ " f'<div class=\"prediction-box success-box\">'\n",
908
+ " f'<h2 style=\"color: #155724;\">✅ High Likelihood of Purchase!</h2>'\n",
909
+ " f'<p style=\"font-size: 18px;\">This customer is <b>likely to purchase</b> the Wellness Tourism Package.</p>'\n",
910
+ " f'<p style=\"font-size: 16px;\">Confidence: <b>{prediction_proba[1]*100:.2f}%</b></p>'\n",
911
+ " f'</div>',\n",
912
+ " unsafe_allow_html=True\n",
913
+ " )\n",
914
+ " st.success(\"💡 **Recommendation:** Prioritize follow-up with this customer!\")\n",
915
+ " else:\n",
916
+ " st.markdown(\n",
917
+ " f'<div class=\"prediction-box warning-box\">'\n",
918
+ " f'<h2 style=\"color: #856404;\">⚠️ Low Likelihood of Purchase</h2>'\n",
919
+ " f'<p style=\"font-size: 18px;\">This customer is <b>unlikely to purchase</b> the Wellness Tourism Package.</p>'\n",
920
+ " f'<p style=\"font-size: 16px;\">Confidence: <b>{prediction_proba[0]*100:.2f}%</b></p>'\n",
921
+ " f'</div>',\n",
922
+ " unsafe_allow_html=True\n",
923
+ " )\n",
924
+ " st.info(\"💡 **Recommendation:** Consider alternative packages or additional engagement strategies.\")\n",
925
+ "\n",
926
+ " # Show probability breakdown\n",
927
+ " col_prob1, col_prob2 = st.columns(2)\n",
928
+ " with col_prob1:\n",
929
+ " st.metric(\"Probability of No Purchase\", f\"{prediction_proba[0]*100:.2f}%\")\n",
930
+ " with col_prob2:\n",
931
+ " st.metric(\"Probability of Purchase\", f\"{prediction_proba[1]*100:.2f}%\")\n",
932
+ "\n",
933
+ " except Exception as e:\n",
934
+ " st.error(f\"Error making prediction: {e}\")\n",
935
+ " else:\n",
936
+ " st.error(\"Model not loaded. Please check the model repository.\")\n",
937
+ "\n",
938
+ "# Footer\n",
939
+ "st.markdown(\"---\")\n",
940
+ "st.markdown(\"\"\"\n",
941
+ "<div style='text-align: center; color: #666; padding: 20px;'>\n",
942
+ " <p>🏢 Visit with Us - Wellness Tourism Package Prediction System</p>\n",
943
+ " <p>Built with ❤️ using Streamlit and XGBoost</p>\n",
944
+ "</div>\n",
945
+ "\"\"\", unsafe_allow_html=True)"
946
+ ]
947
+ },
948
+ {
949
+ "cell_type": "markdown",
950
+ "metadata": {
951
+ "id": "07cYzWcIwTL-"
952
+ },
953
+ "source": [
954
+ "## Dependency Handling"
955
+ ]
956
+ },
957
+ {
958
+ "cell_type": "markdown",
959
+ "metadata": {
960
+ "id": "JEgfHL64jU7o"
961
+ },
962
+ "source": [
963
+ "Please ensure that the dependency handling file is named `requirements.txt`."
964
+ ]
965
+ },
966
+ {
967
+ "cell_type": "code",
968
+ "execution_count": 11,
969
+ "metadata": {
970
+ "colab": {
971
+ "base_uri": "https://colab.research.google.com/"
972
+ },
973
+ "id": "nvdmy7Wd9lda"
974
+ },
975
+ "outputs": [
976
+ {
977
+ "name": "stdout",
978
+ "output_type": "stream",
979
+ "text": [
980
+ "Writing tourism_project/deployment/requirements.txt\n"
981
+ ]
982
+ }
983
+ ],
984
+ "source": [
985
+ "%%writefile tourism_project/deployment/requirements.txt\n",
986
+ "pandas==2.2.2\n",
987
+ "numpy==1.26.4\n",
988
+ "huggingface_hub==0.32.6\n",
989
+ "streamlit==1.43.2\n",
990
+ "joblib==1.5.1\n",
991
+ "scikit-learn==1.6.0\n",
992
+ "xgboost==2.1.4"
993
+ ]
994
+ },
995
+ {
996
+ "cell_type": "markdown",
997
+ "metadata": {
998
+ "id": "V4ynzpKNwWS_"
999
+ },
1000
+ "source": [
1001
+ "# Hosting"
1002
+ ]
1003
+ },
1004
+ {
1005
+ "cell_type": "code",
1006
+ "execution_count": 12,
1007
+ "metadata": {
1008
+ "id": "7p5sBvTg9nCW"
1009
+ },
1010
+ "outputs": [],
1011
+ "source": [
1012
+ "os.makedirs(\"tourism_project/hosting\", exist_ok=True)"
1013
+ ]
1014
+ },
1015
+ {
1016
+ "cell_type": "code",
1017
+ "execution_count": 13,
1018
+ "metadata": {
1019
+ "colab": {
1020
+ "base_uri": "https://colab.research.google.com/"
1021
+ },
1022
+ "id": "QlpgAQoXOtKA"
1023
+ },
1024
+ "outputs": [
1025
+ {
1026
+ "name": "stdout",
1027
+ "output_type": "stream",
1028
+ "text": [
1029
+ "Writing tourism_project/hosting/hosting.py\n"
1030
+ ]
1031
+ }
1032
+ ],
1033
+ "source": [
1034
+ "%%writefile tourism_project/hosting/hosting.py\n",
1035
+ "from huggingface_hub import HfApi\n",
1036
+ "import os\n",
1037
+ "\n",
1038
+ "api = HfApi(token=os.getenv(\"HF_TOKEN\"))\n",
1039
+ "api.upload_folder(\n",
1040
+ " folder_path=\"tourism_project/deployment\",\n",
1041
+ " repo_id=\"ananttripathiak/wellness-tourism-prediction\",\n",
1042
+ " repo_type=\"space\",\n",
1043
+ " path_in_repo=\"\",\n",
1044
+ ")\n",
1045
+ "\n",
1046
+ "print(\"Deployment files successfully uploaded to Hugging Face Space!\")"
1047
+ ]
1048
+ },
1049
+ {
1050
+ "cell_type": "markdown",
1051
+ "metadata": {
1052
+ "id": "PuCgAW2hktli"
1053
+ },
1054
+ "source": [
1055
+ "# MLOps Pipeline with Github Actions Workflow"
1056
+ ]
1057
+ },
1058
+ {
1059
+ "cell_type": "markdown",
1060
+ "metadata": {
1061
+ "id": "L5BZr5i8PKVN"
1062
+ },
1063
+ "source": [
1064
+ "**Note:**\n",
1065
+ "\n",
1066
+ "1. Before running the file below, make sure to add the HF_TOKEN to your GitHub secrets to enable authentication between GitHub and Hugging Face.\n",
1067
+ "2. The below code is for a sample YAML file that can be updated as required to meet the requirements of this project."
1068
+ ]
1069
+ },
1070
+ {
1071
+ "cell_type": "code",
1072
+ "execution_count": 14,
1073
+ "metadata": {
1074
+ "id": "H2kXyghROtKA"
1075
+ },
1076
+ "outputs": [],
1077
+ "source": [
1078
+ "os.makedirs(\"tourism_project/.github/workflows\", exist_ok=True)"
1079
+ ]
1080
+ },
1081
+ {
1082
+ "cell_type": "code",
1083
+ "execution_count": 15,
1084
+ "metadata": {
1085
+ "colab": {
1086
+ "base_uri": "https://colab.research.google.com/"
1087
+ },
1088
+ "id": "M5J4Kq2ROtKA"
1089
+ },
1090
+ "outputs": [
1091
+ {
1092
+ "name": "stdout",
1093
+ "output_type": "stream",
1094
+ "text": [
1095
+ "Writing tourism_project/.github/workflows/pipeline.yml\n"
1096
+ ]
1097
+ }
1098
+ ],
1099
+ "source": [
1100
+ "%%writefile tourism_project/.github/workflows/pipeline.yml\n",
1101
+ "name: Tourism Package Prediction Pipeline\n",
1102
+ "\n",
1103
+ "on:\n",
1104
+ " push:\n",
1105
+ " branches:\n",
1106
+ " - main\n",
1107
+ "\n",
1108
+ "jobs:\n",
1109
+ "\n",
1110
+ " register-dataset:\n",
1111
+ " runs-on: ubuntu-latest\n",
1112
+ " steps:\n",
1113
+ " - uses: actions/checkout@v3\n",
1114
+ " - name: Install Dependencies\n",
1115
+ " run: pip install -r tourism_project/requirements.txt\n",
1116
+ " - name: Upload Dataset to Hugging Face Hub\n",
1117
+ " env:\n",
1118
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1119
+ " run: python tourism_project/model_building/data_register.py\n",
1120
+ "\n",
1121
+ " data-prep:\n",
1122
+ " needs: register-dataset\n",
1123
+ " runs-on: ubuntu-latest\n",
1124
+ " steps:\n",
1125
+ " - uses: actions/checkout@v3\n",
1126
+ " - name: Install Dependencies\n",
1127
+ " run: pip install -r tourism_project/requirements.txt\n",
1128
+ " - name: Run Data Preparation\n",
1129
+ " env:\n",
1130
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1131
+ " run: python tourism_project/model_building/prep.py\n",
1132
+ "\n",
1133
+ " model-traning:\n",
1134
+ " needs: data-prep\n",
1135
+ " runs-on: ubuntu-latest\n",
1136
+ " steps:\n",
1137
+ " - uses: actions/checkout@v3\n",
1138
+ " - name: Install Dependencies\n",
1139
+ " run: pip install -r tourism_project/requirements.txt\n",
1140
+ " - name: Start MLflow Server\n",
1141
+ " run: |\n",
1142
+ " nohup mlflow ui --host 0.0.0.0 --port 5000 &\n",
1143
+ " sleep 5\n",
1144
+ " - name: Model Building\n",
1145
+ " env:\n",
1146
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1147
+ " run: python tourism_project/model_building/train.py\n",
1148
+ "\n",
1149
+ " deploy-hosting:\n",
1150
+ " runs-on: ubuntu-latest\n",
1151
+ " needs: [model-traning,data-prep,register-dataset]\n",
1152
+ " steps:\n",
1153
+ " - uses: actions/checkout@v3\n",
1154
+ " - name: Install Dependencies\n",
1155
+ " run: pip install -r tourism_project/requirements.txt\n",
1156
+ " - name: Push files to Frontend Hugging Face Space\n",
1157
+ " env:\n",
1158
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1159
+ " run: python tourism_project/hosting/hosting.py\n"
1160
+ ]
1161
+ },
1162
+ {
1163
+ "cell_type": "markdown",
1164
+ "metadata": {
1165
+ "id": "J029tYPq4Rmq"
1166
+ },
1167
+ "source": [
1168
+ "```\n",
1169
+ "name: Tourism Project Pipeline\n",
1170
+ "\n",
1171
+ "on:\n",
1172
+ " push:\n",
1173
+ " branches:\n",
1174
+ " - main # Automatically triggers on push to the main branch\n",
1175
+ "\n",
1176
+ "jobs:\n",
1177
+ "\n",
1178
+ " register-dataset:\n",
1179
+ " runs-on: ubuntu-latest\n",
1180
+ " steps:\n",
1181
+ " - uses: actions/checkout@v3\n",
1182
+ " - name: Install Dependencies\n",
1183
+ " run: <add_code_here>\n",
1184
+ " - name: Upload Dataset to Hugging Face Hub\n",
1185
+ " env:\n",
1186
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1187
+ " run: <add_code_here>\n",
1188
+ "\n",
1189
+ " data-prep:\n",
1190
+ " needs: register-dataset\n",
1191
+ " runs-on: ubuntu-latest\n",
1192
+ " steps:\n",
1193
+ " - uses: actions/checkout@v3\n",
1194
+ " - name: Install Dependencies\n",
1195
+ " run: <add_code_here>\n",
1196
+ " - name: Run Data Preparation\n",
1197
+ " env:\n",
1198
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1199
+ " run: <add_code_here>\n",
1200
+ "\n",
1201
+ "\n",
1202
+ " model-traning:\n",
1203
+ " needs: data-prep\n",
1204
+ " runs-on: ubuntu-latest\n",
1205
+ " steps:\n",
1206
+ " - uses: actions/checkout@v3\n",
1207
+ " - name: Install Dependencies\n",
1208
+ " run: <add_code_here>\n",
1209
+ " - name: Start MLflow Server\n",
1210
+ " run: |\n",
1211
+ " nohup mlflow ui --host 0.0.0.0 --port 5000 & # Run MLflow UI in the background\n",
1212
+ " sleep 5 # Wait for a moment to let the server starts\n",
1213
+ " - name: Model Building\n",
1214
+ " env:\n",
1215
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1216
+ " run: <add_code_here>\n",
1217
+ "\n",
1218
+ "\n",
1219
+ " deploy-hosting:\n",
1220
+ " runs-on: ubuntu-latest\n",
1221
+ " needs: [model-traning,data-prep,register-dataset]\n",
1222
+ " steps:\n",
1223
+ " - uses: actions/checkout@v3\n",
1224
+ " - name: Install Dependencies\n",
1225
+ " run: <add_code_here>\n",
1226
+ " - name: Push files to Frontend Hugging Face Space\n",
1227
+ " env:\n",
1228
+ " HF_TOKEN: ${{ secrets.HF_TOKEN }}\n",
1229
+ " run: <add_code_here>\n",
1230
+ "\n",
1231
+ "```"
1232
+ ]
1233
+ },
1234
+ {
1235
+ "cell_type": "markdown",
1236
+ "metadata": {
1237
+ "id": "T9fgZ_Mq3zzp"
1238
+ },
1239
+ "source": [
1240
+ "**Note:** To use this YAML file for our use case, we need to\n",
1241
+ "\n",
1242
+ "1. Go to the GitHub repository for the project\n",
1243
+ "2. Create a folder named ***.github/workflows/***\n",
1244
+ "3. In the above folder, create a file named ***pipeline.yml***\n",
1245
+ "4. Copy and paste the above content for the YAML file into the ***pipeline.yml*** file"
1246
+ ]
1247
+ },
1248
+ {
1249
+ "cell_type": "markdown",
1250
+ "metadata": {
1251
+ "id": "PvEUJ-t5kdxH"
1252
+ },
1253
+ "source": [
1254
+ "## Requirements file for the Github Actions Workflow"
1255
+ ]
1256
+ },
1257
+ {
1258
+ "cell_type": "code",
1259
+ "execution_count": 16,
1260
+ "metadata": {
1261
+ "colab": {
1262
+ "base_uri": "https://colab.research.google.com/"
1263
+ },
1264
+ "id": "nfqWcLRm-dga"
1265
+ },
1266
+ "outputs": [
1267
+ {
1268
+ "name": "stdout",
1269
+ "output_type": "stream",
1270
+ "text": [
1271
+ "Writing tourism_project/requirements.txt\n"
1272
+ ]
1273
+ }
1274
+ ],
1275
+ "source": [
1276
+ "%%writefile tourism_project/requirements.txt\n",
1277
+ "huggingface_hub==0.32.6\n",
1278
+ "datasets==3.6.0\n",
1279
+ "pandas==2.2.2\n",
1280
+ "numpy==1.26.4\n",
1281
+ "scikit-learn==1.6.0\n",
1282
+ "xgboost==2.1.4\n",
1283
+ "mlflow==3.0.1\n",
1284
+ "joblib==1.5.1"
1285
+ ]
1286
+ },
1287
+ {
1288
+ "cell_type": "markdown",
1289
+ "metadata": {
1290
+ "id": "BA6mP-Ebkm3O"
1291
+ },
1292
+ "source": [
1293
+ "## Github Authentication and Push Files"
1294
+ ]
1295
+ },
1296
+ {
1297
+ "cell_type": "markdown",
1298
+ "metadata": {
1299
+ "id": "T84Ei-g9Z2uw"
1300
+ },
1301
+ "source": [
1302
+ "* Before moving forward, we need to generate a secret token to push files directly from Colab to the GitHub repository.\n",
1303
+ "* Please follow the below instructions to create the GitHub token:\n",
1304
+ " - Open your GitHub profile.\n",
1305
+ " - Click on ***Settings***.\n",
1306
+ " - Go to ***Developer Settings***.\n",
1307
+ " - Expand the ***Personal access tokens*** section and select ***Tokens (classic)***.\n",
1308
+ " - Click ***Generate new token***, then choose ***Generate new token (classic)***.\n",
1309
+ " - Add a note and select all required scopes.\n",
1310
+ " - Click ***Generate token***.\n",
1311
+ " - Copy the generated token and store it safely in a notepad."
1312
+ ]
1313
+ },
1314
+ {
1315
+ "cell_type": "code",
1316
+ "execution_count": 17,
1317
+ "metadata": {
1318
+ "colab": {
1319
+ "base_uri": "https://colab.research.google.com/"
1320
+ },
1321
+ "id": "KPDx4gqGh7cO"
1322
+ },
1323
+ "outputs": [
1324
+ {
1325
+ "name": "stdout",
1326
+ "output_type": "stream",
1327
+ "text": [
1328
+ "Reading package lists... Done\n",
1329
+ "Building dependency tree... Done\n",
1330
+ "Reading state information... Done\n",
1331
+ "git is already the newest version (1:2.34.1-1ubuntu1.15).\n",
1332
+ "0 upgraded, 0 newly installed, 0 to remove and 1 not upgraded.\n",
1333
+ "Cloning into 'Tourism_Project'...\n",
1334
+ "remote: Enumerating objects: 99, done.\u001b[K\n",
1335
+ "remote: Counting objects: 100% (99/99), done.\u001b[K\n",
1336
+ "remote: Compressing objects: 100% (89/89), done.\u001b[K\n",
1337
+ "remote: Total 99 (delta 31), reused 0 (delta 0), pack-reused 0 (from 0)\u001b[K\n",
1338
+ "Receiving objects: 100% (99/99), 111.21 KiB | 4.28 MiB/s, done.\n",
1339
+ "Resolving deltas: 100% (31/31), done.\n",
1340
+ "mv: cannot move '/content/tourism_project/' to '/content/Tourism_Project/tourism_project': Directory not empty\n"
1341
+ ]
1342
+ }
1343
+ ],
1344
+ "source": [
1345
+ "# Install Git\n",
1346
+ "!apt-get install git\n",
1347
+ "\n",
1348
+ "# Set your Git identity (replace with your details)\n",
1349
+ "!git config --global user.email \"ananttripathi1996@gmail.com\"\n",
1350
+ "!git config --global user.name \"ananttripathi\"\n",
1351
+ "\n",
1352
+ "# Clone your GitHub repository\n",
1353
+ "!git clone https://github.com/ananttripathi/Tourism_Project.git\n",
1354
+ "\n",
1355
+ "# Move your folder to the repository directory\n",
1356
+ "!mv /content/tourism_project/ /content/Tourism_Project"
1357
+ ]
1358
+ },
1359
+ {
1360
+ "cell_type": "code",
1361
+ "execution_count": null,
1362
+ "metadata": {
1363
+ "colab": {
1364
+ "base_uri": "https://colab.research.google.com/"
1365
+ },
1366
+ "id": "IuUahCwVigon"
1367
+ },
1368
+ "outputs": [
1369
+ {
1370
+ "name": "stdout",
1371
+ "output_type": "stream",
1372
+ "text": [
1373
+ "/content/Tourism_Project\n",
1374
+ "On branch main\n",
1375
+ "Your branch is up to date with 'origin/main'.\n",
1376
+ "\n",
1377
+ "nothing to commit, working tree clean\n",
1378
+ "Everything up-to-date\n"
1379
+ ]
1380
+ }
1381
+ ],
1382
+ "source": [
1383
+ "# Change directory to the cloned repository\n",
1384
+ "%cd Tourism_Project/\n",
1385
+ "\n",
1386
+ "# Add the new folder to Git\n",
1387
+ "!git add .\n",
1388
+ "\n",
1389
+ "# Commit the changes\n",
1390
+ "!git commit -m \"first commit\"\n",
1391
+ "\n",
1392
+ "# Push to GitHub (you'll need your GitHub credentials; use a personal access token if 2FA enabled)\n",
1393
+ "!git push https://ananttripathi:github_pat_11SPxp30Fxq@github.com/ananttripathi/Tourism_Project.git"
1394
+ ]
1395
+ },
1396
+ {
1397
+ "cell_type": "markdown",
1398
+ "metadata": {
1399
+ "id": "v-i8Jdyz-_L1"
1400
+ },
1401
+ "source": [
1402
+ "# Output Evaluation"
1403
+ ]
1404
+ },
1405
+ {
1406
+ "cell_type": "markdown",
1407
+ "metadata": {
1408
+ "id": "FTK8Bpda_UHg"
1409
+ },
1410
+ "source": [
1411
+ "- GitHub (link to repository, screenshot of folder structure and executed workflow)"
1412
+ ]
1413
+ },
1414
+ {
1415
+ "cell_type": "code",
1416
+ "execution_count": 19,
1417
+ "metadata": {
1418
+ "id": "6qzzesaG_Xw8"
1419
+ },
1420
+ "outputs": [],
1421
+ "source": [
1422
+ "# TODO: Add Screenshot After Deployment\n",
1423
+ "#\n",
1424
+ "# After pushing to GitHub and running the workflow, add a screenshot showing:\n",
1425
+ "# 1. GitHub repository structure (showing tourism_project folder and all subfolders)\n",
1426
+ "# 2. GitHub Actions workflow execution (all 4 jobs completed successfully with green checkmarks)\n",
1427
+ "#\n",
1428
+ "# Screenshot should show:\n",
1429
+ "# ✅ register-dataset job completed\n",
1430
+ "# ✅ data-prep job completed\n",
1431
+ "# ✅ model-training job completed\n",
1432
+ "# ✅ deploy-hosting job completed\n",
1433
+ "#\n",
1434
+ "# You can also provide the GitHub repository URL here"
1435
+ ]
1436
+ },
1437
+ {
1438
+ "cell_type": "code",
1439
+ "execution_count": 19,
1440
+ "metadata": {
1441
+ "id": "qXS_wjMdOtKM"
1442
+ },
1443
+ "outputs": [],
1444
+ "source": []
1445
+ },
1446
+ {
1447
+ "cell_type": "markdown",
1448
+ "metadata": {
1449
+ "id": "IidBPl95OtKM"
1450
+ },
1451
+ "source": [
1452
+ "*[Screenshot: GitHub repository structure and folder layout]*"
1453
+ ]
1454
+ },
1455
+ {
1456
+ "cell_type": "markdown",
1457
+ "metadata": {
1458
+ "id": "P8QouFeUOtKM"
1459
+ },
1460
+ "source": [
1461
+ "*[Screenshot: GitHub Actions workflow execution]*"
1462
+ ]
1463
+ },
1464
+ {
1465
+ "cell_type": "markdown",
1466
+ "metadata": {
1467
+ "id": "3KDN31V2_YSr"
1468
+ },
1469
+ "source": [
1470
+ "- Streamlit on Hugging Face (link to HF space, screenshot of Streamlit app)"
1471
+ ]
1472
+ },
1473
+ {
1474
+ "cell_type": "code",
1475
+ "execution_count": 20,
1476
+ "metadata": {
1477
+ "id": "NuIUdj3b_ZYV"
1478
+ },
1479
+ "outputs": [],
1480
+ "source": [
1481
+ "# TODO: Add Screenshot After Deployment\n",
1482
+ "#\n",
1483
+ "# After the workflow completes and app is deployed, add a screenshot showing:\n",
1484
+ "# 1. Live Streamlit app running on Hugging Face Spaces\n",
1485
+ "# 2. Making a sample prediction with the interface\n",
1486
+ "# 3. Prediction results displayed with confidence scores\n",
1487
+ "#\n",
1488
+ "# Also provide the Hugging Face Space URL:\n",
1489
+ "# Expected format: https://huggingface.co/spaces/<your-username>/wellness-tourism-prediction\n",
1490
+ "#\n",
1491
+ "# Screenshot should show:\n",
1492
+ "# ✅ App interface with input forms\n",
1493
+ "# ✅ Prediction button\n",
1494
+ "# ✅ Results with confidence scores\n",
1495
+ "# ✅ Recommendations"
1496
+ ]
1497
+ },
1498
+ {
1499
+ "cell_type": "markdown",
1500
+ "metadata": {
1501
+ "id": "tp06I8AnOtKM"
1502
+ },
1503
+ "source": [
1504
+ "*[Screenshot: Streamlit app on Hugging Face Spaces]*"
1505
+ ]
1506
+ },
1507
+ {
1508
+ "cell_type": "code",
1509
+ "execution_count": 20,
1510
+ "metadata": {
1511
+ "id": "mK_xcFxlOtKM"
1512
+ },
1513
+ "outputs": [],
1514
+ "source": []
1515
+ },
1516
+ {
1517
+ "cell_type": "markdown",
1518
+ "metadata": {
1519
+ "id": "zWAag1sAOtKM"
1520
+ },
1521
+ "source": [
1522
+ "*[Screenshot: Prediction results with confidence scores]*"
1523
+ ]
1524
+ },
1525
+ {
1526
+ "cell_type": "markdown",
1527
+ "metadata": {
1528
+ "id": "fN8j9-3nW8G9"
1529
+ },
1530
+ "source": [
1531
+ "<font size=6 color=\"navyblue\">Power Ahead!</font>\n",
1532
+ "___"
1533
+ ]
1534
+ }
1535
+ ],
1536
+ "metadata": {
1537
+ "kernelspec": {
1538
+ "display_name": "Python [conda env:anaconda3] *",
1539
+ "language": "python",
1540
+ "name": "conda-env-anaconda3-py"
1541
+ },
1542
+ "language_info": {
1543
+ "codemirror_mode": {
1544
+ "name": "ipython",
1545
+ "version": 3
1546
+ },
1547
+ "file_extension": ".py",
1548
+ "mimetype": "text/x-python",
1549
+ "name": "python",
1550
+ "nbconvert_exporter": "python",
1551
+ "pygments_lexer": "ipython3",
1552
+ "version": "3.11.4"
1553
+ }
1554
+ },
1555
+ "nbformat": 4,
1556
+ "nbformat_minor": 0
1557
+ }
UPLOAD_GUIDE.md ADDED
@@ -0,0 +1,223 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # File Upload Guide: Where Each File Goes
2
+
3
+ This guide shows exactly which files are uploaded to which location (Hugging Face or GitHub) and when.
4
+
5
+ ---
6
+
7
+ ## Overview: Three Upload Destinations
8
+
9
+ 1. **Hugging Face Dataset Repo** (`ananttripathiak/engine-maintenance-dataset`)
10
+ 2. **Hugging Face Model Repo** (`ananttripathiak/engine-maintenance-model`)
11
+ 3. **Hugging Face Space** (`ananttripathiak/engine-maintenance-space`)
12
+ 4. **GitHub Repository** (`ananttripathi/engine-predictive-maintenance`)
13
+
14
+ ---
15
+
16
+ ## 1. Hugging Face Dataset Repo
17
+
18
+ **Repo ID**: `ananttripathiak/engine-maintenance-dataset`
19
+ **Created by**: `src/data_register.py` and `src/data_prep.py`
20
+
21
+ ### Files Uploaded:
22
+
23
+ #### A. Raw Data (via `src/data_register.py`)
24
+ - **File**: `data/engine_data.csv`
25
+ - **Uploaded to**: `data/engine_data.csv` in the dataset repo
26
+ - **When**: Run `python src/data_register.py`
27
+
28
+ #### B. Processed Data (via `src/data_prep.py`)
29
+ - **File**: `data/processed/train.csv`
30
+ - **Uploaded to**: `data/train.csv` in the dataset repo
31
+ - **When**: Run `python src/data_prep.py`
32
+
33
+ - **File**: `data/processed/test.csv`
34
+ - **Uploaded to**: `data/test.csv` in the dataset repo
35
+ - **When**: Run `python src/data_prep.py`
36
+
37
+ **Scripts that upload here:**
38
+ - `src/data_register.py` → uploads raw data
39
+ - `src/data_prep.py` → uploads train/test splits
40
+
41
+ ---
42
+
43
+ ## 2. Hugging Face Model Repo
44
+
45
+ **Repo ID**: `ananttripathiak/engine-maintenance-model`
46
+ **Created by**: `src/train.py`
47
+
48
+ ### Files Uploaded:
49
+
50
+ - **File**: `models/best_model.joblib`
51
+ - **Uploaded to**: `model.joblib` in the model repo
52
+ - **When**: Run `python src/train.py` (after training completes)
53
+
54
+ **Scripts that upload here:**
55
+ - `src/train.py` → uploads the trained model
56
+
57
+ ---
58
+
59
+ ## 3. Hugging Face Space (Streamlit App)
60
+
61
+ **Repo ID**: `ananttripathiak/engine-maintenance-space`
62
+ **Created by**: `src/deploy_to_hf.py`
63
+
64
+ ### Files Uploaded:
65
+
66
+ The `src/deploy_to_hf.py` script uploads the entire project folder **except**:
67
+ - `data/` (ignored - too large)
68
+ - `mlruns/` (ignored - MLflow tracking data)
69
+ - `models/` (ignored - model is in model repo)
70
+ - `.github/` (ignored - GitHub-specific)
71
+
72
+ **Files that ARE uploaded to Space:**
73
+ - `src/app.py` ← **Main Streamlit app**
74
+ - `src/inference.py` ← Inference utilities
75
+ - `src/config.py` ← Configuration
76
+ - `Dockerfile` ← Container definition
77
+ - `requirements.txt` ← Python dependencies
78
+ - `README.md` ← Documentation
79
+ - Other `src/*.py` files (if needed by app)
80
+
81
+ **Scripts that upload here:**
82
+ - `src/deploy_to_hf.py` → uploads deployment files
83
+
84
+ ---
85
+
86
+ ## 4. GitHub Repository
87
+
88
+ **Repo URL**: `https://github.com/ananttripathi/engine-predictive-maintenance`
89
+ **Created by**: You (manually via `git push`)
90
+
91
+ ### Files Uploaded:
92
+
93
+ **Everything in the `mlops/` folder**, including:
94
+ - ✅ `data/` (including `engine_data.csv`, `processed/train.csv`, `processed/test.csv`)
95
+ - ✅ `src/` (all Python scripts)
96
+ - ✅ `notebooks/` (EDA notebooks, etc.)
97
+ - ✅ `.github/workflows/pipeline.yml` ← **GitHub Actions workflow**
98
+ - ✅ `requirements.txt`
99
+ - ✅ `Dockerfile`
100
+ - ✅ `README.md`
101
+ - ✅ `models/` (if you want to track model versions in git)
102
+ - ✅ `mlruns/` (MLflow tracking data - optional)
103
+ - ✅ All other project files
104
+
105
+ **How to upload:**
106
+ ```bash
107
+ cd /Users/ananttripathi/Desktop/mlops
108
+ git init
109
+ git add .
110
+ git commit -m "Initial commit: Predictive maintenance MLOps pipeline"
111
+ git remote add origin https://github.com/ananttripathi/engine-predictive-maintenance.git
112
+ git push -u origin main
113
+ ```
114
+
115
+ ---
116
+
117
+ ## Upload Workflow Summary
118
+
119
+ ### Step-by-Step Upload Process:
120
+
121
+ 1. **Data Registration** → Hugging Face Dataset Repo
122
+ ```bash
123
+ python src/data_register.py
124
+ ```
125
+ - Uploads: `data/engine_data.csv` → HF Dataset Repo
126
+
127
+ 2. **Data Preparation** → Hugging Face Dataset Repo
128
+ ```bash
129
+ python src/data_prep.py
130
+ ```
131
+ - Uploads: `data/processed/train.csv` and `test.csv` → HF Dataset Repo
132
+
133
+ 3. **Model Training** → Hugging Face Model Repo
134
+ ```bash
135
+ python src/train.py
136
+ ```
137
+ - Uploads: `models/best_model.joblib` → HF Model Repo
138
+
139
+ 4. **Deploy App** → Hugging Face Space
140
+ ```bash
141
+ python src/deploy_to_hf.py
142
+ ```
143
+ - Uploads: `src/app.py`, `Dockerfile`, `requirements.txt`, etc. → HF Space
144
+
145
+ 5. **Push to GitHub** → GitHub Repository
146
+ ```bash
147
+ git add .
148
+ git commit -m "Complete MLOps pipeline"
149
+ git push origin main
150
+ ```
151
+ - Uploads: Everything → GitHub Repo
152
+
153
+ ---
154
+
155
+ ## What Gets Uploaded Automatically vs Manually
156
+
157
+ ### Automatic (via Scripts):
158
+ - ✅ Hugging Face Dataset Repo → `src/data_register.py` and `src/data_prep.py`
159
+ - ✅ Hugging Face Model Repo → `src/train.py`
160
+ - ✅ Hugging Face Space → `src/deploy_to_hf.py`
161
+ - ✅ GitHub Actions → Runs automatically when you push to GitHub
162
+
163
+ ### Manual:
164
+ - ⚠️ **GitHub Repository** → You need to run `git push` yourself
165
+
166
+ ---
167
+
168
+ ## File Size Considerations
169
+
170
+ ### Large Files (may be ignored):
171
+ - `data/engine_data.csv` → Uploaded to HF Dataset, but you might want to add to `.gitignore` for GitHub
172
+ - `mlruns/` → MLflow tracking data (can be large) - ignored by HF Space deploy
173
+ - `models/best_model.joblib` → Uploaded to HF Model Repo, but you might want to add to `.gitignore` for GitHub
174
+
175
+ ### Recommended `.gitignore`:
176
+ ```
177
+ # Large data files
178
+ data/*.csv
179
+ data/processed/*.csv
180
+
181
+ # MLflow tracking
182
+ mlruns/
183
+
184
+ # Model files (already in HF Model Repo)
185
+ models/*.joblib
186
+
187
+ # Python cache
188
+ __pycache__/
189
+ *.pyc
190
+ .venv/
191
+ ```
192
+
193
+ ---
194
+
195
+ ## Quick Reference Table
196
+
197
+ | File/Folder | HF Dataset | HF Model | HF Space | GitHub |
198
+ |------------|------------|----------|----------|--------|
199
+ | `data/engine_data.csv` | ✅ | ❌ | ❌ | ⚠️ Optional |
200
+ | `data/processed/train.csv` | ✅ | ❌ | ❌ | ⚠️ Optional |
201
+ | `data/processed/test.csv` | ✅ | ❌ | ❌ | ⚠️ Optional |
202
+ | `models/best_model.joblib` | ❌ | ✅ | ❌ | ⚠️ Optional |
203
+ | `src/app.py` | ❌ | ❌ | ✅ | ✅ |
204
+ | `src/train.py` | ❌ | ❌ | ❌ | ✅ |
205
+ | `src/data_prep.py` | ❌ | ❌ | ❌ | ✅ |
206
+ | `Dockerfile` | ❌ | ❌ | ✅ | ✅ |
207
+ | `requirements.txt` | ❌ | ❌ | ✅ | ✅ |
208
+ | `.github/workflows/pipeline.yml` | ❌ | ❌ | ❌ | ✅ |
209
+ | `README.md` | ❌ | ❌ | ✅ | ✅ |
210
+
211
+ **Legend:**
212
+ - ✅ = Uploaded automatically or should be uploaded
213
+ - ❌ = Not uploaded to this location
214
+ - ⚠️ Optional = Can be uploaded but might want to exclude from GitHub due to size
215
+
216
+ ---
217
+
218
+ ## Need Help?
219
+
220
+ - **Hugging Face Dataset**: Check `src/hf_data_utils.py`
221
+ - **Hugging Face Model**: Check `src/hf_model_utils.py`
222
+ - **Hugging Face Space**: Check `src/deploy_to_hf.py`
223
+ - **GitHub**: Standard git commands
USERNAME_SUMMARY.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Username Configuration Summary
2
+
3
+ ## ✅ Updated Configuration
4
+
5
+ All files have been updated with the correct usernames:
6
+
7
+ - **Hugging Face Username**: `ananttripathiak` (with "ak" at the end)
8
+ - **GitHub Username**: `ananttripathi` (no "ak")
9
+
10
+ ## Files Updated
11
+
12
+ ### 1. `src/config.py` ✅
13
+ - **HF_DATASET_REPO**: `ananttripathiak/engine-maintenance-dataset`
14
+ - **HF_MODEL_REPO**: `ananttripathiak/engine-maintenance-model`
15
+ - **HF_SPACE_REPO**: `ananttripathiak/engine-maintenance-space`
16
+
17
+ ### 2. `README.md` ✅
18
+ - Updated all Hugging Face repo examples to use `ananttripathiak`
19
+ - GitHub repository URLs use `ananttripathi`
20
+
21
+ ### 3. `CONFIGURATION_GUIDE.md` ✅
22
+ - Updated examples to show `ananttripathiak` for HF repos
23
+
24
+ ### 4. `GITHUB_REPO_INFO.md` ✅
25
+ - Updated GitHub secrets examples to use `ananttripathiak` for HF repos
26
+ - GitHub URLs use `ananttripathi`
27
+
28
+ ## What You Need to Do
29
+
30
+ ### 1. GitHub Secrets (in your GitHub repo settings)
31
+
32
+ When you add secrets to your GitHub repository, use these exact values:
33
+
34
+ - **`HF_TOKEN`**: Your Hugging Face access token
35
+ - **`HF_DATASET_REPO`**: `ananttripathiak/engine-maintenance-dataset`
36
+ - **`HF_MODEL_REPO`**: `ananttripathiak/engine-maintenance-model`
37
+ - **`HF_SPACE_REPO`**: `ananttripathiak/engine-maintenance-space`
38
+
39
+ ### 2. Local Environment Variables (optional)
40
+
41
+ If you want to override config.py, set these:
42
+
43
+ ```bash
44
+ export HF_TOKEN="hf_your_token_here"
45
+ export HF_DATASET_REPO="ananttripathiak/engine-maintenance-dataset"
46
+ export HF_MODEL_REPO="ananttripathiak/engine-maintenance-model"
47
+ export HF_SPACE_REPO="ananttripathiak/engine-maintenance-space"
48
+ ```
49
+
50
+ ### 3. GitHub Repository URLs
51
+
52
+ Your GitHub repository will be:
53
+ - **Repository**: `https://github.com/ananttripathi/engine-predictive-maintenance`
54
+ - **Actions**: `https://github.com/ananttripathi/engine-predictive-maintenance/actions`
55
+
56
+ ### 4. Hugging Face Repos (will be created automatically)
57
+
58
+ After running the scripts, these will be created:
59
+ - **Dataset**: `https://huggingface.co/datasets/ananttripathiak/engine-maintenance-dataset`
60
+ - **Model**: `https://huggingface.co/ananttripathiak/engine-maintenance-model`
61
+ - **Space**: `https://huggingface.co/spaces/ananttripathiak/engine-maintenance-space`
62
+
63
+ ## Verification
64
+
65
+ All configuration files are now correctly set. You can proceed with:
66
+ 1. Running the scripts locally
67
+ 2. Pushing to GitHub
68
+ 3. Setting up GitHub secrets
69
+ 4. Running the GitHub Actions workflow
WHAT_TO_UPLOAD.md ADDED
@@ -0,0 +1,265 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # What to Upload: GitHub vs Hugging Face
2
+
3
+ ## Quick Summary
4
+
5
+ | Destination | What to Upload | How |
6
+ |------------|----------------|-----|
7
+ | **GitHub** | **Everything** (entire project folder) | `git push` |
8
+ | **HF Dataset** | Raw data + train/test splits | Scripts auto-upload |
9
+ | **HF Model** | Trained model file | Scripts auto-upload |
10
+ | **HF Space** | App files (code, Dockerfile, requirements) | Scripts auto-upload |
11
+
12
+ ---
13
+
14
+ ## 📦 GitHub Repository
15
+
16
+ ### Upload: **Everything in the `mlops/` folder**
17
+
18
+ **What to include:**
19
+ ```
20
+ mlops/
21
+ ├── data/
22
+ │ ├── engine_data.csv ✅ Upload
23
+ │ └── processed/
24
+ │ ├── train.csv ✅ Upload
25
+ │ └── test.csv ✅ Upload
26
+ ├── src/
27
+ │ ├── app.py ✅ Upload
28
+ │ ├── config.py ✅ Upload
29
+ │ ├── data_prep.py ✅ Upload
30
+ │ ├── data_register.py ✅ Upload
31
+ │ ├── train.py ✅ Upload
32
+ │ ├── inference.py ✅ Upload
33
+ │ ├── deploy_to_hf.py ✅ Upload
34
+ │ ├── eda.py ✅ Upload
35
+ │ ├── hf_data_utils.py ✅ Upload
36
+ │ └── hf_model_utils.py ✅ Upload
37
+ ├── notebooks/ ✅ Upload (if you have EDA notebooks)
38
+ ├── models/
39
+ │ └── best_model.joblib ⚠️ Optional (large file)
40
+ ├── .github/
41
+ │ └── workflows/
42
+ │ └── pipeline.yml ✅ Upload (IMPORTANT!)
43
+ ├── requirements.txt ✅ Upload
44
+ ├── Dockerfile ✅ Upload
45
+ ├── README.md ✅ Upload
46
+ └── *.md files ✅ Upload (documentation)
47
+
48
+ ❌ DON'T upload:
49
+ ├── .venv/ ❌ Skip (virtual environment)
50
+ ├── __pycache__/ ❌ Skip (Python cache)
51
+ ├── mlruns/ ❌ Skip (MLflow tracking - can be large)
52
+ └── .git/ ❌ Skip (git metadata)
53
+ ```
54
+
55
+ ### How to Upload to GitHub:
56
+
57
+ ```bash
58
+ cd /Users/ananttripathi/Desktop/mlops
59
+
60
+ # Initialize git (if not already done)
61
+ git init
62
+
63
+ # Create .gitignore to exclude large/unnecessary files
64
+ cat > .gitignore << EOF
65
+ .venv/
66
+ __pycache__/
67
+ *.pyc
68
+ *.pyo
69
+ *.pyd
70
+ .Python
71
+ mlruns/
72
+ *.log
73
+ .DS_Store
74
+ EOF
75
+
76
+ # Add all files
77
+ git add .
78
+
79
+ # Commit
80
+ git commit -m "Initial commit: Predictive Maintenance MLOps Pipeline"
81
+
82
+ # Add remote (replace with your repo URL)
83
+ git remote add origin https://github.com/ananttripathi/engine-predictive-maintenance.git
84
+
85
+ # Push
86
+ git push -u origin main
87
+ ```
88
+
89
+ ---
90
+
91
+ ## 🤗 Hugging Face - Dataset Repo
92
+
93
+ **Repo**: `ananttripathiak/engine-maintenance-dataset`
94
+
95
+ ### Upload: **Data files only**
96
+
97
+ **Files uploaded automatically by scripts:**
98
+ - `data/engine_data.csv` → Uploaded as `data/engine_data.csv`
99
+ - `data/processed/train.csv` → Uploaded as `data/train.csv`
100
+ - `data/processed/test.csv` → Uploaded as `data/test.csv`
101
+
102
+ ### How to Upload:
103
+
104
+ **Option 1: Run the scripts (automatic)**
105
+ ```bash
106
+ # Step 1: Register raw data
107
+ python src/data_register.py
108
+
109
+ # Step 2: Prepare and upload train/test
110
+ python src/data_prep.py
111
+ ```
112
+
113
+ **Option 2: Manual upload via HF Hub**
114
+ - Go to https://huggingface.co/datasets/ananttripathiak/engine-maintenance-dataset
115
+ - Click "Add file" → Upload `data/engine_data.csv`
116
+ - Upload `data/processed/train.csv` and `test.csv`
117
+
118
+ ---
119
+
120
+ ## 🤗 Hugging Face - Model Repo
121
+
122
+ **Repo**: `ananttripathiak/engine-maintenance-model`
123
+
124
+ ### Upload: **Trained model file only**
125
+
126
+ **File uploaded automatically:**
127
+ - `models/best_model.joblib` → Uploaded as `model.joblib`
128
+
129
+ ### How to Upload:
130
+
131
+ **Option 1: Run the training script (automatic)**
132
+ ```bash
133
+ python src/train.py
134
+ # This will:
135
+ # 1. Train the model
136
+ # 2. Save to models/best_model.joblib
137
+ # 3. Upload to HF Model Repo automatically
138
+ ```
139
+
140
+ **Option 2: Manual upload via HF Hub**
141
+ - Go to https://huggingface.co/ananttripathiak/engine-maintenance-model
142
+ - Click "Add file" → Upload `models/best_model.joblib`
143
+ - Rename it to `model.joblib` in the repo
144
+
145
+ ---
146
+
147
+ ## 🤗 Hugging Face - Space (Streamlit App)
148
+
149
+ **Repo**: `ananttripathiak/engine-maintenance-space`
150
+
151
+ ### Upload: **App deployment files**
152
+
153
+ **Files uploaded automatically by script:**
154
+ ```
155
+ ✅ src/app.py (Main Streamlit app)
156
+ ✅ src/inference.py (Inference utilities)
157
+ ✅ src/config.py (Configuration)
158
+ ✅ Dockerfile (Container definition)
159
+ ✅ requirements.txt (Dependencies)
160
+ ✅ README.md (Documentation)
161
+ ✅ Other src/*.py files (If needed by app)
162
+
163
+ ❌ NOT uploaded (ignored):
164
+ ├── data/ ❌ Too large
165
+ ├── mlruns/ ❌ MLflow tracking
166
+ ├── models/ ❌ Model is in Model Repo
167
+ ├── .github/ ❌ GitHub-specific
168
+ └── .venv/ ❌ Virtual environment
169
+ ```
170
+
171
+ ### How to Upload:
172
+
173
+ **Option 1: Run the deployment script (automatic)**
174
+ ```bash
175
+ python src/deploy_to_hf.py
176
+ # This will:
177
+ # 1. Create/update the HF Space
178
+ # 2. Upload all deployment files
179
+ # 3. Configure it as a Streamlit app
180
+ ```
181
+
182
+ **Option 2: Manual upload via HF Hub**
183
+ - Go to https://huggingface.co/spaces/ananttripathiak/engine-maintenance-space
184
+ - Upload files one by one or use HF CLI
185
+
186
+ ---
187
+
188
+ ## 📋 Complete Upload Checklist
189
+
190
+ ### ✅ Step 1: GitHub (Manual - Do First)
191
+ - [ ] Create GitHub repository
192
+ - [ ] Add `.gitignore` file
193
+ - [ ] Push entire `mlops/` folder to GitHub
194
+ - [ ] Add GitHub Secrets (HF_TOKEN, HF_DATASET_REPO, HF_MODEL_REPO, HF_SPACE_REPO)
195
+
196
+ ### ✅ Step 2: Hugging Face Dataset (Automatic)
197
+ - [ ] Set `HF_TOKEN` environment variable
198
+ - [ ] Run `python src/data_register.py` (uploads raw data)
199
+ - [ ] Run `python src/data_prep.py` (uploads train/test)
200
+
201
+ ### ✅ Step 3: Hugging Face Model (Automatic)
202
+ - [ ] Run `python src/train.py` (trains model and uploads to HF)
203
+
204
+ ### ✅ Step 4: Hugging Face Space (Automatic)
205
+ - [ ] Run `python src/deploy_to_hf.py` (deploys app to HF Space)
206
+
207
+ ### ✅ Step 5: Verify
208
+ - [ ] Check GitHub repo: https://github.com/ananttripathi/engine-predictive-maintenance
209
+ - [ ] Check HF Dataset: https://huggingface.co/datasets/ananttripathiak/engine-maintenance-dataset
210
+ - [ ] Check HF Model: https://huggingface.co/ananttripathiak/engine-maintenance-model
211
+ - [ ] Check HF Space: https://huggingface.co/spaces/ananttripathiak/engine-maintenance-space
212
+
213
+ ---
214
+
215
+ ## 🎯 Key Differences
216
+
217
+ | Aspect | GitHub | Hugging Face |
218
+ |--------|--------|--------------|
219
+ | **Purpose** | Code repository & version control | Data/Model/App hosting |
220
+ | **What** | Entire project (code, data, docs) | Specific artifacts (data/model/app) |
221
+ | **How** | Manual `git push` | Automatic via scripts |
222
+ | **Size** | Can be large (includes everything) | Optimized (only what's needed) |
223
+ | **Access** | Public/Private repo | Public datasets/models/spaces |
224
+ | **CI/CD** | GitHub Actions workflow | HF Spaces auto-deploy |
225
+
226
+ ---
227
+
228
+ ## 💡 Pro Tips
229
+
230
+ 1. **GitHub First**: Always push to GitHub first, then run HF scripts
231
+ 2. **Use .gitignore**: Exclude large files like `mlruns/` and `.venv/` from GitHub
232
+ 3. **HF Scripts are Smart**: They automatically create repos if they don't exist
233
+ 4. **Check File Sizes**: HF has file size limits, so scripts exclude large files
234
+ 5. **GitHub Secrets**: Store HF credentials in GitHub Secrets for CI/CD
235
+
236
+ ---
237
+
238
+ ## 🚨 Common Mistakes to Avoid
239
+
240
+ ❌ **Don't upload to GitHub:**
241
+ - Virtual environment (`.venv/`)
242
+ - Large MLflow runs (`mlruns/`)
243
+ - Python cache files (`__pycache__/`)
244
+
245
+ ❌ **Don't upload to HF Space:**
246
+ - Raw data files (too large)
247
+ - Model files (use Model Repo instead)
248
+ - MLflow tracking data
249
+
250
+ ✅ **Do upload to GitHub:**
251
+ - All source code
252
+ - Configuration files
253
+ - Documentation
254
+ - GitHub Actions workflow
255
+
256
+ ✅ **Do upload to HF:**
257
+ - Only what each repo type needs (data → Dataset, model → Model, app → Space)
258
+
259
+ ---
260
+
261
+ ## 📞 Need Help?
262
+
263
+ - **GitHub Issues**: Check your repo settings and secrets
264
+ - **HF Upload Errors**: Verify `HF_TOKEN` is set correctly
265
+ - **File Size Issues**: Check HF file size limits (usually 10GB for datasets)
requirements.txt ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ numpy
2
+ pandas
3
+ scikit-learn
4
+ mlflow
5
+ huggingface_hub
6
+ streamlit
7
+ joblib
8
+ matplotlib
9
+ seaborn
10
+ plotly
11
+