https://github.com/shivamr021/student-admission-predictor
A machine learning app built from scratch using logistic regression. It predicts student admission chances based on test scores. Includes a deployed Streamlit app for real-time predictions.
https://github.com/shivamr021/student-admission-predictor
classification ml ml-project python streamlit-webapp student-admission-predictor
Last synced: 16 days ago
JSON representation
A machine learning app built from scratch using logistic regression. It predicts student admission chances based on test scores. Includes a deployed Streamlit app for real-time predictions.
- Host: GitHub
- URL: https://github.com/shivamr021/student-admission-predictor
- Owner: shivamr021
- License: mit
- Created: 2025-07-18T11:41:50.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2025-07-18T12:10:50.000Z (11 months ago)
- Last Synced: 2025-07-18T15:46:44.252Z (11 months ago)
- Topics: classification, ml, ml-project, python, streamlit-webapp, student-admission-predictor
- Language: Python
- Homepage: https://student-admission-predictor-minimal-version.streamlit.app/
- Size: 20.5 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 🎓 Student Admission Predictor - Logistic Regression (Streamlit App)
Predict your chances of graduate school admission using a simple machine learning model!
This project implements logistic regression from scratch (without using libraries like `sklearn.linear_model`) and wraps it in a clean and interactive **Streamlit web app**.
## 🔗 Live Demo
▶️ [Try the App](https://student-admission-predictor-minimal-version.streamlit.app/)
---
## 📌 Features
- Logistic Regression implemented **from scratch in NumPy**
- User-friendly **Streamlit interface** for predictions
- Predicts **chances of admission (0 or 1)** based on:
- GRE Score
- TOEFL Score
- University Rating
- SOP Strength
- LOR Strength
- CGPA
- Research Experience
---
## 📊 Dataset
- **Source:** [Admission Prediction Dataset](https://www.kaggle.com/datasets/mohansacharya/graduate-admissions)
- **Total Samples:** 500 students
- **Label Used:** `Chance of Admit ≥ 0.75 → Admitted (1)`, otherwise `Not Admitted (0)`
---
## 🧠 Model Highlights
- Custom **sigmoid**, **loss**, **gradient**, and **training loop**
- Uses **min-max normalization**
- Evaluation metrics:
- Accuracy: 87.20%
- Precision: 84.68%
- Recall: 86.24%
- F1 Score: 85.45%
---
## 🚀 How to Run Locally
1. Clone the repository:
```bash
git clone https://github.com/shivamr021/student-admission-predictor.git
cd student-admission-predictor
````
2. Install required packages:
```bash
pip install -r requirements.txt
```
3. Run the Streamlit app:
```bash
streamlit run streamlit_app.py
```
---
## 📁 File Structure
```bash
.
├── streamlit_app.py # Streamlit app (interactive UI)
├── admission_data.csv # Dataset (used for training)
├── README.md # This file
├── requirements.txt # Dependencies
```
---
## 💡 Learning Outcome
* Implemented end-to-end **binary classification** without using machine learning libraries
* Learned to build a basic **ML app with Streamlit**
* Understood core ML evaluation metrics
* Practiced data preprocessing and normalization
---
## 🌐 Tech Stack
* Python
* NumPy
* Pandas
* Matplotlib
* Streamlit
---
## ✨ Future Improvements
* Add model trained using `scikit-learn` for comparison
* Allow file upload to test batch data
* Deploy advanced model on platforms like HuggingFace or Render
---
## 👤 Author
**Shivam Rathod**
📬 [LinkedIn](https://linkedin.com/in/shivamrathod021)
---
## 📜 License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.