Commits

Breadcrumbs

History for

oncudnn

User selector

All users

All time

Commit History

Commits on Jul 16, 2016

• Add script to automatically put the Kaldi libraries we link with in the right order; use it to modify the Makefiles. Minor top-level Makefile fix.

  danpovey

  committed
  dbae7fa

  Copy full SHA for dbae7fa

Commits on May 20, 2016

• Add missing dependencies to Makefiles

  akreal

  committed
  5b3fccd

  Copy full SHA for 5b3fccd

Commits on Apr 6, 2016

• arpa2fst: count the bos/eos warninig towards the limit

  kkm000

  committed
  62d3e17

  Copy full SHA for 62d3e17

Commits on Apr 5, 2016

• Merge pull request #634 from kkm000/arpa-3

  Show description for 2fd63cb
  danpovey

  committed
  2fd63cb

  Copy full SHA for 2fd63cb

• Kill arpa2fst use_log10 option (#640)

  

  kkm

  committed
  e3c9155

  Copy full SHA for e3c9155

• Merge pull request #649 from kkm000/arpa-diags

  Show description for fea62e6
  danpovey

  committed
  fea62e6

  Copy full SHA for fea62e6

• Improved ARPA compilers diagnostics.

  Show description for 60bdba9
  

  kkm

  committed
  60bdba9

  Copy full SHA for 60bdba9

Commits on Apr 4, 2016

• fix to bug in arpa2fst regarding zero n-gram counts.

  danpovey

  committed
  d03abac

  Copy full SHA for d03abac

Commits on Apr 3, 2016

• Add option --ilabel-sort to arpa2fst

  

  kkm

  committed
  33b6d6f

  Copy full SHA for 33b6d6f

• Allow anything before \data\ (#634, comments)

  

  kkm

  committed
  c77451d

  Copy full SHA for c77451d

Commits on Mar 27, 2016

• small issue in the arpa2fst code

  jtrmal

  committed
  6353fed

  Copy full SHA for 6353fed

Commits on Mar 23, 2016

• Merge pull request #579 from kkm000/arpa-2

  Show description for 8fab002
  danpovey

  committed
  8fab002

  Copy full SHA for 8fab002

• Removed the --natural-base option from the two ARPA compilers.

  

  kkm

  committed
  a6f3b82

  Copy full SHA for a6f3b82

Commits on Mar 17, 2016

• Fixed data truncation bug in history key

  

  kkm

  committed
  f0f6dd4

  Copy full SHA for f0f6dd4

Commits on Mar 16, 2016

• Added code-level support for the ',bg' option on sequential table readers, causing them to read the data in a background thread. Helps speed up GPU-based nnet training, for instance.

  danpovey

  committed
  99a39ff

  Copy full SHA for 99a39ff

• changed std::cerr to KALDI_LOG

  

  xiaohui-zhang

  authored and

  danpovey

  committed
  bf08599

  Copy full SHA for bf08599

• specifying namespace std when using cerr in src/lm/kaldi-lmtable.cc. This only affects compilation with some versions of openfst.

  

  xiaohui-zhang

  authored and

  danpovey

  committed
  fcd8a52

  Copy full SHA for fcd8a52

Commits on Mar 15, 2016

• Implemented correct optimized key selection

  Show description for e144c8b
  

  kkm

  committed
  e144c8b

  Copy full SHA for e144c8b

• Added command switches to arpa2fst utility

  

  kkm

  committed
  fb25a77

  Copy full SHA for fb25a77

• Added epsilon-substitution mode tests

  Show description for fa0171d
  

  kkm

  committed
  fa0171d

  Copy full SHA for fa0171d

Commits on Mar 11, 2016

• Major rewrite of the ARPA compiler

  

  kkm

  committed
  f57dc92

  Copy full SHA for f57dc92

Commits on Mar 9, 2016

• changed std::cerr to KALDI_LOG

  xiaohui-zhang

  committed
  59ded21

  Copy full SHA for 59ded21

• specifying namespace std when using cerr in src/lm/kaldi-lmtable.cc. This only affects compilation with some versions of openfst.

  xiaohui-zhang

  committed
  d343505

  Copy full SHA for d343505

Commits on Jan 29, 2016

• Merge remote-tracking branch 'upstream/master' into chain

  danpovey

  committed
  ad7951d

  Copy full SHA for ad7951d

Commits on Jan 28, 2016

• remove redundant type hints

  funous

  committed
  f587473

  Copy full SHA for f587473

Commits on Jan 26, 2016

• Fix compilation issues when using KALDI_DOUBLEPRECISION=1 caused by hardcoded types

  funous

  committed
  a422890

  Copy full SHA for a422890

• change to arpa-file-parser.cc to suppress spurious compiler warning

  danpovey

  committed
  e46406b

  Copy full SHA for e46406b

Commits on Jan 22, 2016

• Changes per @danpovey's review in #458

  

  kkm

  committed
  1aab0b6

  Copy full SHA for 1aab0b6

• Improved error messages for ARPA file parsing

  Show description for 6265183
  

  kkm

  committed
  6265183

  Copy full SHA for 6265183

Commits on Jan 19, 2016

• Separated ARPA parsing from const LM construction

  

  kkm

  committed
  e0cbd32

  Copy full SHA for e0cbd32

Commits on Dec 1, 2015

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

  chenguoguo

  committed
  f0798f4

  Copy full SHA for f0798f4

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

  chenguoguo

  committed
  f10bc7e

  Copy full SHA for 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

  committed
  7166563

  Copy full SHA for 7166563

Commits on Nov 26, 2015

• modify arpa-reading code. generate error when n-grams speficied in header and statistics not match

  

  Minhua Wu

  committed
  a4283db

  Copy full SHA for a4283db

Commits on Oct 20, 2015

• Removing null-pointer checks as mentioned in #89

  jtrmal

  committed
  da07306

  Copy full SHA for da07306