Overview

This project is an end-to-end machine learning solution for predicting student placements. It involves data collection, preprocessing, model training, hyperparameter tuning, and deployment using Streamlit.

Project Structure

├── data
│   ├── raw_data
│   │   ├── raw.csv
│   ├── cleaned_data
│   ├── preprocess_data
├── models
│   ├── best_model
│   ├── model
│   ├── scaler
├── notebooks
│   ├── data_collection.ipynb
│   ├── data_cleaning.ipynb
│   ├── preprocessing.ipynb
│   ├── model_train.ipynb
│   ├── hyperparameter_tuning.ipynb
├── src
│   ├── data_collection.py
│   ├── data_cleaning.py
│   ├── preprocessing.py
│   ├── model_train.py
│   ├── hyperparameter_tuning.py
│   ├── app.py
│   ├── main.py
│   ├── streamlit.py
│   ├── test.py
│   ├── requirements.txt
├── placementdata.csv
├── .gitignore

Pipeline Steps

1. Data Collection

• The raw dataset (raw.csv) is gathered and stored in the data/raw_data/ directory.
• The data_collection.ipynb and data_collection.py scripts are used to fetch and save the data.

2. Data Cleaning

• Missing values, duplicate records, and irrelevant features are handled.
• The cleaned data is stored in the data/cleaned_data/ directory.
• Implemented in data_cleaning.ipynb and data_cleaning.py.

3. Data Preprocessing

• Feature engineering, encoding categorical variables, and feature scaling.
• Preprocessed data is saved in data/preprocess_data/.
• Implemented in preprocessing.ipynb and preprocessing.py.

4. Model Training

• Multiple classification models were trained:
  • Logistic Regression: 79.8%
  • Decision Tree Classifier: 71.25%
  • Random Forest Classifier: 79.1%
  • AdaBoost Classifier: 81.05% (Best Model)
  • Gradient Boosting Classifier: 80.3%
• Implemented in model_train.ipynb and model_train.py.
• Trained models are saved in the models/model/ directory.

5. Hyperparameter Tuning

• Models are optimized to improve accuracy.
• The best-performing model (AdaBoost) is saved in models/best_model/.
• Implemented in hyperparameter_tuning.ipynb and hyperparameter_tuning.py.

6. Model Deployment

• The best model is used for predictions.
• A Streamlit web application is created for user interaction.
• Implemented in app.py and streamlit.py.

7. Testing and Evaluation

• The final model is evaluated on test data.
• Results and metrics are logged.
• Implemented in test.py.

Installation and Usage

Requirements

Install dependencies using:

pip install -r requirements.txt

Running the App

Run the Streamlit app with:

streamlit run app.py

Future Improvements

• Improve model accuracy with additional feature engineering.
• Integrate a database for real-time student placement tracking.
• Expand deployment options using Flask/Django for a web-based solution.

Author

Prakshi Karkera