Numeric Character Recognition (NCR)

The project is based on the Kaggle competition, Numeric Character Recognition, which involves taking images of handwritten single digits from the MNIST dataset and determining the digit in each image using classification algorithms. Please visit the competition page for the full details and competition rules.

Table of Contents

• Project Overview
• Installation
• Running the Notebook
• Running on Google Colab
• Citation

Project Overview

The objective is to correctly identify digits from the MNIST ("Modified National Institute of Standards and Technology") dataset, which contains handwritten images, using a Convolutional Neural Network (CNN) to achieve the highest categorization accuracy, or the proportion of test images that are correctly classified.

File Descriptions

1. ncr_final_version.ipynb

This Jupyter notebook contains the implementation of the CNN for classifying MNIST handwritten digits. It includes data preprocessing, model training, evaluation, and visualization of results.

2. ncr/

This directory contains training and testing datasets and the submission file for the competition. It includes:

• train.csv: Gray-scale training images of hand-drawn digits with labels (0-9). Each image is 28 x 28 pixels, with pixel values ranging from 0 (light) to 255 (dark).
• test.csv: Same as the training set, but without the label column
• sample_submission.csv: Contains 28,000 lines, one for each image in the test set, with two columns: ImageId and Label (predicted digit)

3. final_results.csv

This file contains the predicted digits for each test image (test.csv), following the format of sample_submission.csv.

Installation

1. Clone the repository to your local machine:

git clone https://github.com/masheharashid/numeric_character_recognition.git
cd numeric_character_recognition

Running the Notebook

1. Launch the Jupyter Notebook Interface

• Open the terminal and navigate to the project directory. Then, launch the Jupyter Notebook interface and open the Python notebook file

  jupyter notebook ncr_final_version.ipynb

2. Follow the instructions in the notebook and run each cell in order

Running on Google Colab

Option 1:

• Click on the "Open in Colab" button at the top of the ncr_final_version.ipynb file or go to this link

Option 2:

1. Upload the Notebook

• Open Google Colab
• Upload the ncr_final_version.ipynb file

2. Mount the Google Drive (Optional)

• If the NCR files are stored in Google Drive, mount the drive to access them:

  from google.colab import drive
  drive.mount('/content/drive')

3. Run the Notebook

• Execute the cells in order as instructed in the notebook

Citation

Akaash Tripathi. (2024). Numeric Character Recognition (NCR). Kaggle. https://kaggle.com/competitions/ncr