The project aimed to implement Deep NN / RNN based solution in order to develop flexible methods that are able to adaptively fillin, backfill, and predict time-series using a large number of heterogeneous training datasets.

Using advanced deep learning techniques on the MNIST dataset. Over 98% validation set accuracy.

Implemented Adam optimizer in python

Gesture Recognition by CNN created using Networks Library created by me.

Library which can be used to build feed forward NN, Convolutional Nets, Linear Regression, and Logistic Regression Models.

Generate novel artistic images using neural style transfer algorithm

3-layer linear neural network to classify the MNIST dataset using the TensorFlow

Google Street View House Number(SVHN) Dataset, and classifying them through CNN

Implemenation of DDPG with numpy only (without Tensorflow)

Deep Neural Network implementation for recognizing digit using MNIST data set and Tensorflow

Aibrite Machine Learning Framework

Fully connected neural network for digit classification using MNIST data

Fully connected network on MNIST data using Tensorflow

The Deep Learning exercises provided in DataCamp

Creation of a MultiLayer Perceptron using Back Propagation Algorithm. It was trained to efficiently classify the data into two sets:exit and stay. This was able to predict whether a customer might stay with the bank or leave it in future.

Image inpainting of Depth Images using Deep Image Prior

An AI agent for Pocket Tanks

A repository for comparison of Adam and L-BFGS-B methods using coco benchmark.

Deep Neural Networks like Single Layer Perceptron and Multi Layer Perceptron implementation using Tensorflow library on Datasets like MNIST and Naval Mine for categorical Classification. Saving and Restoring Tensorflow "Variables" weights for testing.