fix: incorrect folder name for openblas installation #4873
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…to generate .counts file (kaldi-asr#4324) rnnlm/get_unigram_counts.sh does not exist. We normally use rnnlm/ensure_counts_present.sh to generate the '.counts' files
Refer to the README.md to each eg directory for description.
* Software distributed on GitHub is downloaded from GitHub:
- cub (official): rework rules to avoid excessive re-downloads,
and change tarball format to .tar.gz to match other software.
- SCTK (official)
- sph2pipe (mirror at burrmill, stable)
* Fix build of sph2pipe. Some tarballs of the same version 2.5
did not accept the patch from extras/, and make -j N for large N
stopped on a prompt from patch(1) that has scrolled far up,
creating an impression that make had hung. The currently used
mirror does not require a patch, and cleanly builds with its own
Makefile. Rework build rules to use the Makefile, and remove
the patch.
* Retire 'openblas' rule: it no longer invokes install_openblas.sh
and directs the user to use the script instead. The rule was
incorrect, because it ignored OPENBLAS_VERSION manifestly set in
the Makefile
* Remove a patch for SCTK: its only purpose had been to add missing
#include's and declarations, which generated warnings in C code.
Instead, suppress all compilation warnings during build.
Fixes: kaldi-asr#4311
Fixes: kaldi-asr#4323
* Fork off tools/.gitignore from the root file. * Make gitignores sensible by using wildcards. Listing every possible versioned package name had been unmaintainable. * Make all extra scripts executable, and trim EOL/EOF whitespace.
* chain2 small fix: no groups-per-minibatch option * replace groups_per_minibatch with minibatch_size and recover steps/nnet3/chain2/train.sh
Co-authored-by: Srikanth MADIKERI <[email protected]>
…] (kaldi-asr#4350) Ubuntu 20.04+ packages an unknown 20.x MKL version. The installation in /opt/intel is still preferred, but if none were found or supplied by the user, the package is probed as well, the last in order of probing. Closes kaldi-asr#4262
…fo constructor, fixes kaldi-asr#4379 (kaldi-asr#4380)
* Updated FP16 and TF32 modes using cublasGemmEx to move away from deprecated features * Changed cu-device to disable default TF32 and added a cmdline flag for it instead. This should make it so that inference workflow is unaffected. * Fixed some test failures
* egs/wsj/s5/steps/nnet3/chain2/train.sh: gpu_cmd_opt set based on use_gpu value * egs/mini_librispeech/s5/local/chain2/tuning/run_tdnn_1a.sh: train.sh called with train_cmd instead of cuda_cmd * wsj/s5/steps/nnet3/chain2/train.sh: syntax bug in final.mdl check fixed Co-authored-by: Srikanth MADIKERI <[email protected]>
* Also my editor has removed some trailing whitespaces.
The example for the post-to-tacc fails , but with the correct of `ark:- |` there is no piping error
Coauthored-By: Jonghwan Hyeon <[email protected]>
Co-authored-by: npovey <[email protected]>
…ion_dependent_subword_lexicon.py (kaldi-asr#4794)
* Update run_blstm.sh fix bug aspire run_blstm.sh * Update egs/aspire/s5/local/nnet3/run_blstm.sh Co-authored-by: Cy 'kkm' Katsnelson <[email protected]> Co-authored-by: Cy 'kkm' Katsnelson <[email protected]>
* Remove unused variable. * cudadecoder: Make word alignment optional. For CTC models using word pieces or graphemes, there is not enough positional information to use the word alignment. I tried marking every unit as "singleton" word_boundary.txt, but this explodes the state space very, very often. See: nvidia-riva/riva-asrlib-decoder#3 With the "_" character in CTC models predicting word pieces, we at the very least know which word pieces begin a word and which ones are either in the middle of the word or the end of a word, but the algorithm would still need to be rewritten, especially since "blank" is not a silence phoneme (it can appear between). I did look into using the lexicon-based word alignment. I don't have a specific complaint about it, but I did get a weird error where it couldn't create a final state at all in the output lattice, which caused Connect() to output an empty lattice. This may be because I wasn't quite sure how to handle the blank token. I treat it as its own phoneme, bcause of limitations in TransitionInformation, but this doesn't really make any sense. Needless to say, while the CTM outputs of the cuda decoder will be correct from a WER point of view, their time stamps won't be correct, but they probably never were in the first place, for CTC models.
Fix "glossaries_opt" variable name at line number 39. It's misspelled due to which words in the glossaries weren't reserved while creating BPE.
This is to fix a CI error. It appears that this is from using "ubuntu-latest" in the CI workflow. It got upgraded to ubuntu 22.04 automatically, and this doesn't have python2.7 by default.
…aldi-asr#4820) - the audio data no longer exist in that path - the beamformer config was missing in 'ami/s5b', it's taken from 'ami/s5'
Do not use ADL to invoke std::binary_search
SRILM: allow bypassing download/extraction during automated installation
Fix -Wdeprecated-copy from c++11
…#4838) function is not used anyway
This continues the work started in kaldi-asr#4809.
* Fix matrix data offset for large matrices * Fix overflow in cudamatrix too
Fix download location in install_liblbfgs.sh
|
I think this is resolved in master now, via #4881 sorry for late response. If specifically needed in 5.4 branch you could submit patch with just that change. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.