Skip to content

Created a four layer Neural Network that works on the MNIST Fashion dataset.

Notifications You must be signed in to change notification settings

RohilH/Deep-Learning-Neural-Network

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Deep-Learning-Neural-Network

Created a four layer Neural Network that works on the MNIST Fashion dataset. This code builds a 4-layer neural network with 256 hidden nodes per layer except the last layer which has 10 (number of classes) layers. I use Minibatch Gradient Descent, which runs for a given number of epochs and does the following per epoch: Shuffle the training data Split the data into batches (use batch size of 200) For each batch (subset of data): feed batch into the 4-layer neural network Compute loss and update weights Observe the total loss and go to next iteration

The code can be run by running python nn_main.py

About

Created a four layer Neural Network that works on the MNIST Fashion dataset.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages