Deep Learning sample programs using PyTorch in C++

Real time face landmarking using decision trees and NN autoencoders

C++ implementation of neural networks library with Keras-like API. Contains majority of commonly used layers, losses and optimizers. Supports sequential and multi-input-output (flow) models. Supports single CPU, Multi-CPU and GPU tensor operations (using cuDNN and cuBLAS).

RLE, Huffman, JPEG

A Variational Autoencoder, written entirely in C++

Naive implementation of the generative adversarial network (GAN) training written in c++ using mxnet cpp API

Autoencoders

Comparison of multiple methods for calculating MNIST hand-written digits similarity.

3 part project: A. bottleneck autoencoder, B. manhattan distance, C. earth mover's distance

C++ Deep Learning Library (DLFS-TUM)

Reducing MNIST image data dimensionality by extracting the latent space representations of an Autoencoder model. Comparing these latent space representations to the default MNIST representation

bilingual word embeddings mapping using fastText

Implementation of clustering algorithms and optimizations in C++. Benchmarked on the MNIST handwritten digit dataset

Use AutoEncoders to facilitate indexing of high dimensional data (C++, LibTorch)

Autoencoder dimensionality reduction, EMD-Manhattan metrics comparison and classifier based clustering on MNIST dataset.

Algorithmic problem-solving project using autoencoders for dimension reduction, nearest neighbor search algorithms, and K-means clustering. Developed as part of a university course on software development for algorithmic problems.