A component of the DiffEq ecosystem for enabling sensitivity analysis for scientific machine learning (SciML). Optimize-then-discretize, discretize-then-optimize, adjoint methods, and more for ODEs, SDEs, DDEs, DAEs, etc.

A trainable convolutional neural network inside a fragment shader

An implementation of various metaheuristics adapted to train neural networks

For Azimuth ACT course

A multi-layer, feedforward neural network (FNN) implementation in Java

Testing various examples and code for Machine Learning using TensorFlow

Assignments for the Machine Learning course (COL774) at IITD

Backpropogation algorithm implemented in Python 3

This gradient decent library has not only a determination function, but it has emotions too, as an add on! (Currently the emotions are anger, surprise and excitement)(By emotions, I JUST MEAN BUILT-IN FUNCTIONS FOR EFFICIENT TRAINING)

The Deep Learning exercises provided in DataCamp

Creating a basic NN from scratch to detect numbers from the MNIST dataset using Numpy

Multivariate Classification Using a Feed-Forward Neural Network and Backpropagation.

All the assignments of DLFA course IIT KGP

Implementation of Logistic Regression considering a single Neural Network Node

Implementation of a simple neural network with one hidden layer and training with backpropagation

Pytorch/ without Pytorch implementation of mnist handwriting dataset

Built from scratch neural network library using only NumPy

This Jupyter Notebook contains a separate class of Neural Network.