PA2:
- ML_PA2.ipynb: It contains implementation of LeNet-5 model which is done manually using Tensorflow and Numpy. After training it over 15 epochs with Optimizer: Adam the accuracy 95% approx. for MNIST dataset
- Report file: PDF and word file contains all the results chart and explanation of the results of accuracies.
- archive: This folder contains the MNIST dataset which is downloaded from (https://www.kaggle.com/datasets/hojjatk/mnist-dataset). All files in ubyte format on Kaggle there was an implementation I found to retreive images from these ubyte files. Total 60000 training images and 10000 test images of 0-9 (28x28) digits