An end-to-end ML project that predicts the chances of getting an admit in a University based on different features like University Rating, Student’s Undergrad GPA, GRE & TOEFL scores, Research experience and the quality of SOP & LOR. It returns the chance of getting an admit in a particular university in percentage format.
Following algorithms were used:
- Linear Regression
- Artificial Neural Network (ANN)
- Random Forest
- Decision Tree
Linear Regression had the highest accuracy among all the algorithms. Various Regression KPIs like Root Mean Square Error (RMSE), Mean Square Error (MSE), Mean Absolute Error (MSE), R-square (r2_score) were analysed. In the end Linear Regression model was deployed due to lack of storage on Heroku. All other models can run locally.
- tensorflow
pip install tensorflow - scikit-learn
pip install -U scikit-learn - flask
pip install flask - pickle
pip install pickle - matplotlib.pyplot
pip install matplotlib - seaborn
pip install seaborn - numpy
pip install numpy - pandas
pip install pandas
-
Fork this repository
-
Clone this GitHub Repository in your system.
-
In command line (anaconda prompt), go to the folder that contains all the project files. Run the command
python app.pywhich will give you an address likelocalhost:5000orlocalhost:8050Copy and paste it in the address bar of web browser. -
The project's interface will load locally on the web browser.
Link: GradAdmits App