https://github.com/ayushshahh/fespn

A neural network made to predict final exam scores of students
https://github.com/ayushshahh/fespn

mlp mlp-regressor multilayer-perceptron neural-network prediction-model scikit-learn

Last synced: about 2 months ago
JSON representation

A neural network made to predict final exam scores of students

• Host: GitHub
• URL: https://github.com/ayushshahh/fespn
• Owner: AyushShahh
• License: mit
• Created: 2023-07-20T19:03:02.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2023-08-11T12:22:44.000Z (almost 3 years ago)
• Last Synced: 2025-02-02T11:42:47.824Z (over 1 year ago)
• Topics: mlp, mlp-regressor, multilayer-perceptron, neural-network, prediction-model, scikit-learn
• Language: Python
• Homepage: https://passornot.netlify.app/
• Size: 192 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# fespn

A neural network made to predict final exam scores of students.


Read the detailed blog on [Medium](https://ayushah.medium.com/i-trained-a-neural-network-to-predict-final-exam-scores-of-my-classmates-dab1d979a5c4)


Repo of frontend implementation is [here](https://github.com/AyushShahh/passornot)

There are three files:

- mlp.py - For training the neural network and predicting the scores

- finalcal.py - For estimating the previous final exam scores based on the grade and internals

- internals.py - For calculating the internal marks for current semester

A neural network with ReLU activation function and Stochastic Gradient Descent as an optimizer (solver) has been used. It has two hidden layers with 5 and 4 nodes respectively.

![Neural Network diagram](assets/nn.png)

### students_data.csv

name,roll no 1,roll no 2,phymse1,phymse2,bemse1,bemse2,mat1mse1,mat1mse2,bmemse1,bmemse2,esmse1,esmse2,mat2mse1,mat2mse2,beemse1,beemse2,ppsmse1,ppsmse2,engmse1,engmse2,egdmse1,egdmse2,attendance,full

### final_data.csv

name,phy_i,be_i,mat_i,bme_i,es_i,phy_g,be_g,mat_g,bme_g,es_g

This work is licensed under the [MIT License](/LICENSE). View the license file for details.