Commits

Branch selector

 master

User selector

All users

All time

Commit History

Commits on Feb 10, 2020

• Merge pull request #303 from NTT123/fix-batch-size-1

  Show description for 2583315
  rafaelvalle

  authored
  2583315

  Copy full SHA for 2583315

• Merge pull request #304 from NTT123/remove-tensorboardX

  Show description for ca5a22a
  rafaelvalle

  authored
  ca5a22a

  Copy full SHA for ca5a22a

• remove tensorboardX; use torch.utils.tensorboard

  

  ntt123

  committed
  438b939

  Copy full SHA for 438b939

• fix error when batch size = 1

  

  ntt123

  committed
  14dbc37

  Copy full SHA for 14dbc37

Commits on Jan 9, 2020

• train.py: printing correct variable

  

  Rafael Valle

  committed
  6d0635e

  Copy full SHA for 6d0635e

Commits on Dec 12, 2019

• utils.py: compatibility with new pytorch

  

  Rafael Valle

  committed
  a513db5

  Copy full SHA for a513db5

• logger.py: compatibility with new tensorboardX

  

  Rafael Valle

  committed
  37a033d

  Copy full SHA for 37a033d

Commits on Oct 26, 2019

• train.py: reporting the right variable

  rafaelvalle

  authored
  70d37f9

  Copy full SHA for 70d37f9

Commits on Apr 22, 2019

• Merge pull request #188 from jybaek/fixed-waveglow-link

  Show description for 131c146
  rafaelvalle

  authored
  131c146

  Copy full SHA for 131c146

Commits on Apr 19, 2019

• Fixed link to download waveglow from inference.py

  jybaek

  committed
  d5321ff

  Copy full SHA for d5321ff

Commits on Apr 3, 2019

• README.md: clarifying terminology

  

  rafaelvalle

  committed
  c76ac3b

  Copy full SHA for c76ac3b

• README.md: using proper nomenclature

  

  rafaelvalle

  committed
  e3d2d0a

  Copy full SHA for e3d2d0a

• README.md: updating terminology

  

  rafaelvalle

  committed
  a992aea

  Copy full SHA for a992aea

• Merge branch 'master' of https://github.com/NVIDIA/tacotron2

  

  rafaelvalle

  committed
  eb2a171

  Copy full SHA for eb2a171

• README.md: adding instructions to install apex

  

  rafaelvalle

  committed
  821bfeb

  Copy full SHA for 821bfeb

• Dockerfile: updating to use latest pytorch and apex

  

  rafaelvalle

  committed
  d6670c8

  Copy full SHA for d6670c8

• train.py: using amp for mixed precision training

  

  rafaelvalle

  committed
  0274619

  Copy full SHA for 0274619

• inference.ipynb: adding fp16 inference

  

  rafaelvalle

  committed
  bb20035

  Copy full SHA for bb20035

• model.py: renaming variables, removing dropout from lstm cell state, removing conversions now handled by amp

  

  rafaelvalle

  committed
  1480f82

  Copy full SHA for 1480f82

• logger.py: using new pytorch api

  

  rafaelvalle

  committed
  087c867

  Copy full SHA for 087c867

Commits on Mar 19, 2019

• train.py: changing dataloder params given sampler

  rafaelvalle

  committed
  ece7d3f

  Copy full SHA for ece7d3f

Commits on Mar 16, 2019

• train.py: shuffling at every epoch

  

  rafaelvalle

  committed
  f37998c

  Copy full SHA for f37998c

• README.md: adding explanation on training from pre-trained model

  

  rafaelvalle

  committed
  bff304f

  Copy full SHA for bff304f

• train.py: adding routine to warm start and ignore layers, e.g. embedding.weight

  

  rafaelvalle

  committed
  3869781

  Copy full SHA for 3869781

• hparams.py: adding ignore_layers argument to ignore text embedding layers when warm_starting

  

  rafaelvalle

  committed
  bb67613

  Copy full SHA for bb67613

Commits on Mar 15, 2019

• inference.ipynb: adding code to remove waveglows bias

  

  rafaelvalle

  committed
  af1f71a

  Copy full SHA for af1f71a

• stft.py: moving window_sum to cuda if magnitude is cuda

  

  rafaelvalle

  committed
  fc0d34c

  Copy full SHA for fc0d34c

Commits on Feb 1, 2019

• Merge pull request #136 from GrzegorzKarchNV/master

  Show description for f2c94d9
  rafaelvalle

  authored
  f2c94d9

  Copy full SHA for f2c94d9

• Fixing concatenation error for fp16 ditributed training

  grzegorz-k-karch

  committed
  df4a466

  Copy full SHA for df4a466

Commits on Dec 9, 2018

• inference.ipynb: reverting fp16 inference for now

  

  rafaelvalle

  committed
  825ffa4

  Copy full SHA for 825ffa4

Commits on Dec 6, 2018

• inference.ipynb: changing waverglow inference fo fp16

  

  rafaelvalle

  committed
  4d7b041

  Copy full SHA for 4d7b041

Commits on Nov 28, 2018

• train.py: val logger on gpu 0 only

  

  rafaelvalle

  committed
  6e43055

  Copy full SHA for 6e43055

• hparams.py: distributed using tcp

  

  rafaelvalle

  committed
  3973b3e

  Copy full SHA for 3973b3e

• distributed.py: replacing to avoid distributed error

  

  rafaelvalle

  committed
  52a30bb

  Copy full SHA for 52a30bb

• train.py: renaming variable to n_gpus

  

  rafaelvalle

  committed
  0ad65cc

  Copy full SHA for 0ad65cc

Pagination

Previous

Next