https://github.com/mohammed-majid/image-classification-nn
Image classification using custom Neural Network (No ML packages)
https://github.com/mohammed-majid/image-classification-nn
custom multiclass-classification neural-network numpy
Last synced: 2 months ago
JSON representation
Image classification using custom Neural Network (No ML packages)
- Host: GitHub
- URL: https://github.com/mohammed-majid/image-classification-nn
- Owner: Mohammed-Majid
- Created: 2024-07-13T20:06:58.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2024-08-31T04:53:30.000Z (almost 2 years ago)
- Last Synced: 2025-01-17T05:08:36.906Z (over 1 year ago)
- Topics: custom, multiclass-classification, neural-network, numpy
- Language: Jupyter Notebook
- Homepage:
- Size: 11 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# MNIST Digit Recognition using Neural Network
-------
This repository contains code for training and evaluating a neural network from scratch (no ML packages) to recognize handwritten digits from the MNIST dataset.
------
## Overview
The main script model.ipynb demonstrates the process of:
- Pre-processing: Loading the MNIST dataset, normalizing the data, and preparing it for training and testing.
- Neural Network: Initializing the parameters, implementing forward and backward propagation, and updating the parameters.
- Training: Training the neural network on the MNIST training set.
- Evaluation: Assessing the model's performance on both training and test data, including accuracy and visualization of learning curves.
- Prediction: Visualizing the predictions of the neural network on sample images.
-------
## Usage
- Clone the Repository
- Install Dependencies: Make sure you have the necessary dependencies installed. You can install them using pip:
```
pip install numpy pandas matplotlib
```
- Run the Script: Execute the script to train the model and evaluate its performance.
-------
## Requirements
- Python 3.x
- numpy
- pandas
- matplotlib
--------
## Project Structure
- mnist_nn.py: The file that contains the script for loading data, defining the neural network, training, and evaluation.
- train-images.idx3-ubyte: The dataset file containing training images.
- train-labels.idx1-ubyte: The dataset file containing training labels.
- t10k-images.idx3-ubyte: The dataset file containing test images.
- t10k-labels.idx1-ubyte: The dataset file containing test labels.