Commits

Branch selector

 master

User selector

chenguoguo

Datepicker

All time

Commit History

Commits on Jun 30, 2021

• [egs] GigaSpeech recipe (#4586)

  chenguoguo

  committedJun 30, 2021
  170a1fc

Commits on Jan 6, 2016

• Changing CuVector to CuVectorBase in CuSparseMatrix::CopyElementsToVec

  chenguoguo

  committedJan 6, 2016
  bcc5b50

Commits on Jan 5, 2016

• 1. Changed CopyToVec in CuSparseMatrix/SparseMatrix to CopyElementsToVec, and removed OtherReal version. Note that I didn't remove the OtherReal version of CopyElementsToVec in SparseVector as that…

  chenguoguo

  committedJan 5, 2016
  e8ffc5f

• Moving CuSparseMatrix/SparseMatrix out of CuVector/Vector

  chenguoguo

  committedJan 5, 2016
  00a3100

Commits on Dec 22, 2015

• Fixing compatibility issue with OS X in utils/prepare_lang.sh. Thanks Xavier Anguera for pointing this out.

  chenguoguo

  committedDec 22, 2015
  898121c

Commits on Dec 15, 2015

• WARNING: this commit may change decoding results slightly due to different pruning effects. Added fstpushspecial to LG stage of graph building to make LG.fst more stochastic. Tested on WSJ, it incr…

  chenguoguo

  committedDec 15, 2015
  a456ef3

Commits on Dec 1, 2015

• Removed ReadIntegerArray() and WriteIntegerArray() from lm/const-arpa-lm.cc

  chenguoguo

  committedDec 1, 2015
  f0798f4

• Moved ReadIntegerArray() and WriteIntegerArray() to const-arpa-lm.cc; changed ReadInternalDeprecated() to ReadInternalOldFormat()

  chenguoguo

  committedDec 1, 2015
  f10bc7e

Commits on Nov 28, 2015

• WIP: documented ConstArpaLm stuff in const-arpa-lm.h; fixed I/O bug (was using 25% more of what's needed on the disk); changed lm_states_size_ to int64

  chenguoguo

  committedNov 28, 2015
  7166563

Commits on Oct 14, 2015

• Cleaning up rnnlm related stuff in WSJ; also adding rnnlm lattice rescoring and updating RESULTS

  chenguoguo

  committedOct 14, 2015
  f90f07c

Commits on Oct 5, 2015

• Fixed strict-aliasing issue using union in const ARPA related code; Also fixed some style issue that I saw in the related code

  chenguoguo

  committedOct 5, 2015
  efc0598

• update documentation: add a reference in the example of using a different vocabulary in online decoding

  chenguoguo

  committedOct 5, 2015
  3509b5a

• update documentation: modify the example of using a different vocabulary in online decoding

  chenguoguo

  committedOct 5, 2015
  c4d0cf9

• update documentation: adding examples of using a different vocabulary in online decoding

  chenguoguo

  committedOct 5, 2015
  45a7d28

Commits on Oct 4, 2015

• update documentation: adding examples of using user-specific language model in online decoding

  chenguoguo

  committedOct 4, 2015
  d53dd36

Commits on Oct 2, 2015

• addressing comments for mikolov-rnnlm-lib.{h,cc}, also fixing various style issues in those two files

  chenguoguo

  committedOct 2, 2015
  5ffe7cc

Commits on Oct 1, 2015

• modifying copyright text for code modified from Tomas Mikolov's RNNLM toolkit

  chenguoguo

  committedOct 1, 2015
  c531678

• addressing comments from reviewers for rnnlm lattice rescoring

  chenguoguo

  committedOct 1, 2015
  41c143d

Commits on Sep 23, 2015

• Merge pull request #1 from hainan-xv/rnnlm

  chenguoguo

  committedSep 23, 2015
  4db306b

Commits on Sep 9, 2015

• Adding initial code for rnnlm lattice rescoring (with Mikolov's tool).

  chenguoguo

  committedSep 9, 2015
  6042649

Commits on Sep 2, 2015

• Fixing the segfault issue in lattice-oracle.cc: used lattice composition instead of GetOracleLattice(); also fixed various style issues that I noticed.

  chenguoguo

  committedSep 2, 2015
  f569d0c

Commits on Aug 7, 2015

• Adding some unit test for CPU version of SparseMatrix; various style fixes;

  chenguoguo

  committedAug 7, 2015
  ecbc27a

• Adding unit test for CuSparseMatrix; fixing various issues/bugs in both CPU and GPU version

  chenguoguo

  committedAug 7, 2015
  90fd69c

Commits on Aug 5, 2015

• trunk: small bug fix in babel scripts

  chenguoguo

  committedAug 5, 2015
  6684585

Commits on Aug 3, 2015

• Adding Frobenius norm for sparse matrix

  chenguoguo

  committedAug 3, 2015
  7a90beb

Commits on Aug 2, 2015

• Adding more functions to sparse matrix; Fixing small bugs in nnet3; nnet3 now compiles again. Code has not been tested yet, will add unit test tomorrow

  chenguoguo

  committedAug 2, 2015
  65a0da8

Commits on Jul 30, 2015

• intermedia work on sparse matrix, nnet3bin now compiles again; working on more functions as well as unit test

  chenguoguo

  committedJul 30, 2015
  2e9f6f5

Commits on Jul 7, 2015

• trunk: adding optional 3rd argument to lattice-to-ctm-conf so that it can also reads the word sequence from input

  chenguoguo

  committedJul 7, 2015
  638fd70

Commits on Jun 30, 2015

• trunk: fixing OTWV calculation in kws_oracle_threshold.pl

  chenguoguo

  committedJun 30, 2015
  f161fca

Commits on Jun 29, 2015

• trunk: adding higher order LM support in segmentation scripts

  chenguoguo

  committedJun 29, 2015
  b1560b5

Commits on Jun 28, 2015

• trunk: removing a duplicate line introduced in Nagendra's last commit

  chenguoguo

  committedJun 28, 2015
  5ff8cd8

Commits on Jun 23, 2015

• trunk: fixing arpa format LM data section reading in const-arpa-lm.cc; adding some more warnings

  chenguoguo

  committedJun 23, 2015
  066a610

Commits on Jun 18, 2015

• trunk: adding LogSoftmax component to nnet2 setup

  chenguoguo

  committedJun 18, 2015
  1afc4b3

Commits on Jun 17, 2015

• trunk: fixes to lm reading code regarding <s> and </s>

  chenguoguo

  committedJun 17, 2015
  6581d03

Commits on Jun 16, 2015

• trunk: adding --keep-ali option to rnnlm rescoring; adding support for G.carpa in rnnlm rescoring; plus some small fixes

  chenguoguo

  committedJun 16, 2015
  bb63ec4

Pagination

PreviousNext