https://github.com/matin-ghorbani/mlp_from_scratch

Implement a multi layer perceptron from scratch
https://github.com/matin-ghorbani/mlp_from_scratch

classification from-scratch mlp multilayer-perceptron

Last synced: 9 days ago
JSON representation

Implement a multi layer perceptron from scratch

• Host: GitHub
• URL: https://github.com/matin-ghorbani/mlp_from_scratch
• Owner: matin-ghorbani
• License: mit
• Created: 2024-05-12T12:37:35.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-05-12T13:03:39.000Z (over 1 year ago)
• Last Synced: 2025-04-04T00:23:05.810Z (7 months ago)
• Topics: classification, from-scratch, mlp, multilayer-perceptron
• Language: Jupyter Notebook
• Homepage:
• Size: 101 KB
• Stars: 6
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# MLP (Multilayer Perceptron) From Scratch

### Install Dependencies:

```bash

pip install -r requirements.txt

```

## I implemented my neural network as a class:

### Losses and Accuracies of my MLP for 100 epochs on Mnist dataset:

![Losses and Accuracies of my MLP](./results/check_losses_accuracies_model.png)

## Information About It:

- **accuracy_train: *0.9916492693110647***

- **accuracy_test: *0.8888888888888888***

- **loss_train: *0.0024207044180093113***

- **loss_test: *0.015910050921797165***