MultiLayer Perceptron from Scratch in MATLAB for image classification

Emanuel Maziero

Telecommunications Engineering, last semester.

Simple MLP from scratch in MATLAB for the image classification task using the MNIST handwritten digits dataset.

Download the "mnist_train.csv" and "mnist_test.csv" files;
Ensure that all .m files are in the same folder;
Move the .csv files to the folder where the .m files are located;
Run MNIST_train.m and check if the files b_1, b_2, b_out, w_1, w_2, and w_out were created (figures will be generated from running this script);
Run MNIST_val.m, which will use the files created in step 4.

Name		Name	Last commit message	Last commit date
Latest commit History 13 Commits
MNIST_train.m		MNIST_train.m
MNIST_val.m		MNIST_val.m
README.md		README.md
acuracia.m		acuracia.m
backprop.m		backprop.m
cross_entropy.m		cross_entropy.m
dcross_entropy.m		dcross_entropy.m
dsigmoid.m		dsigmoid.m
forprop.m		forprop.m
sigmoid.m		sigmoid.m
softmax.m		softmax.m