Skip to content

Latest commit

 

History

History
35 lines (23 loc) · 1.29 KB

File metadata and controls

35 lines (23 loc) · 1.29 KB

Handwritten Digit Recognition on Persian(Farsi) MNIST Dataset

In this project i implemented MLP network using pytorch and train it on Perian MNIST dataset.

🔨 Libraries used

🔑 Prerequisites

All the dependencies and required libraries are included in the file requirements.txt see here

📈 Dataset

I uses persian digit handwritten dataset like MNIST. You can find here

🔑 Neural Network

MLP network is used for this implementation. find MLP class in this file: mlp.py

💡 Working

After cloning the repository, run mlp_classification_letters.py on your IDE or open terminal and write following command:

$ python3 mlp_classification_letters.py

💾 Results

You can see result of thid cod in following picture:

image

License

This project is licensed under the MIT License.