MNIST

Project Summary: MNIST Digit Classification

The MNIST digit classification project aimed to develop and deploy a machine learning model capable of accurately classifying handwritten digits from the MNIST dataset. The MNIST dataset consists of 28x28 pixel grayscale images of handwritten digits (0 through 9), making it a standard benchmark for image classification tasks.

QuickStart

• Colab: Easy Keras setup notebook available at link

Python/Keras Folder:

#Install miniconda3 (windows)
curl https://repo.anaconda.com/miniconda/Miniconda3-latest-Windows-x86_64.exe -o miniconda.exe
start /wait "" miniconda.exe /S
del miniconda.exe

#Create environment with the necessary libraries
conda create --name <env> --file requirement.txt

#Run K_CNN.py or K_NN.py

Webcam Integration

• From Python/Webcam run main.py

#Include the model.keras within the folder and load your prediction model

model = tf.keras.models.load_model('MNIST_CNN_model.keras')