A simple processor that takes packets of data (excluding preamble) from a DCC Packet Analyzer (such as the NCE Packet Analyzer) and returns objects that represent the different messages.

Evaluated the word vectors learned from both nce and cross entropy loss functions using word analogy tests

A comprehensive ist of all courses I have taken and successfully completed in University 🎓, Coursera 📗, Udemy etc.

Train utilities(DCC/NCE/LocoNet/LCC)

InfoNCE, Soft Nearest Neighbour on MNIST and Fashion-MNIST

Flow Contrastive Estimation (FCE) PyTorch Implementation on 2D data

Compute package engines constraint to the most restrictive range version used by your dependencies.

NCE Learning

📈🎓

a pytorch version lstm language model

language models toolkits with hierarchical softmax setting

Noise Contrastive Estimation (NCE) in PyTorch

Re-implementation of the Noise Contrastive Estimation algorithm for pyTorch, following "Noise-contrastive estimation: A new estimation principle for unnormalized statistical models." (Gutmann and Hyvarinen, AISTATS 2010)

Tensorflow NCE loss in Keras

Tensorflow implementation of "Representation Learning with Contrastive Predictive Coding"

The Noise Contrastive Estimation for softmax output written in Pytorch

Unsupervised Feature Learning via Non-parametric Instance Discrimination