Commits

Commits on Nov 13, 2021

Let's actually try to finish this thing
norabelrose
committedNov 13, 2021

Commits on Sep 26, 2021

Ungodly number of changes
norabelrose
committedSep 26, 2021

Commits on May 29, 2021

Committing to save TSNE code I'm about to delete
norabelrose
committedMay 29, 2021

Commits on May 2, 2021

Preparing to make this the new master
norabelrose
committedMay 2, 2021

Commits on Apr 27, 2021

This configuration seems to legit work
norabelrose
committedApr 27, 2021

Commits on Apr 21, 2021

Sparse Transformer VAEs working on Wikipedia
norabelrose
committedApr 21, 2021

Commits on Apr 7, 2021

Purely transformer VAE now working!
norabelrose
committedApr 7, 2021

Commits on Apr 4, 2021

Progress reproducing "Lagging Inference Networks"
norabelrose
committedApr 4, 2021

Commits on Apr 2, 2021

Factored __main__.py into different scripts
norabelrose
committedApr 2, 2021

Commits on Mar 31, 2021

More removal of unnecessary code / refactoring
norabelrose
committedMar 31, 2021
Last commit before ckpt BC-breaking changes
norabelrose
committedMar 31, 2021
Some more fixes
norabelrose
committedMar 31, 2021

Commits on Mar 30, 2021

Code functional again after big rename
norabelrose
committedMar 30, 2021
Versioning untracked files
norabelrose
committedMar 30, 2021
Add all the files I realized weren't included in previous 2 commits
norabelrose
committedMar 30, 2021
Small fixes
norabelrose
committedMar 30, 2021
Switched to snake case format for file names
norabelrose
committedMar 30, 2021
Refactoring, code cleanup
norabelrose
committedMar 30, 2021

Commits on Mar 28, 2021

Removed old unused code
norabelrose
committedMar 28, 2021

Commits on Mar 25, 2021

A bunch of updates I forget
norabelrose
committedMar 25, 2021

Commits on Mar 15, 2021

VQVAE model now training successfully
norabelrose
committedMar 15, 2021

Commits on Mar 8, 2021

Non-autoregressive DAAE now training successfully
norabelrose
committedMar 8, 2021

Commits on Feb 28, 2021

QuantizedHierarchicalVAE now working
norabelrose
committedFeb 28, 2021

Commits on Feb 27, 2021

Added adversarially regularized autoencoder
norabelrose
committedFeb 27, 2021

Commits on Feb 24, 2021

ContinuousHierarchicalVAE's decoder no longer 'cheats' by getting direct access to encoder states
norabelrose
committedFeb 24, 2021

Commits on Feb 19, 2021

Refactoring HierarchicalVAE into Quantized and Continuous
norabelrose
committedFeb 19, 2021

Commits on Feb 18, 2021

Refactored some VQ-VAE code into Quantizer class
norabelrose
committedFeb 18, 2021
An absurd amount of new models and refactoring
norabelrose
committedFeb 18, 2021

Commits on Jan 31, 2021

Added support for RNN decoder heads to HierarchicalAutoencoder
norabelrose
committedJan 31, 2021

Commits on Jan 28, 2021

Added KLAnnealing callback and LSTMLanguageModel benchmark
norabelrose
committedJan 28, 2021

Commits on Jan 27, 2021

Unconditional sampling now seems to work
norabelrose
committedJan 27, 2021

Commits on Jan 26, 2021

Refactored HierarchicalAutoencoder and FunnelTransformer to use hidden_state_coroutine()
norabelrose
committedJan 26, 2021

Commits on Jan 23, 2021

Implemented AggressiveEncoderTraining callback; LSTMAutoencoder now trains successfully
norabelrose
committedJan 23, 2021

Commits on Jan 22, 2021

Adding LSTM VAE model as benchmark
norabelrose
committedJan 22, 2021

Commits on Jan 19, 2021

Now passes backward compatibility unit test
norabelrose
committedJan 19, 2021