Welcome to
mirror list
, hosted at
ThFree Co
, Russian Federation.
github.com/marian-nmt/marian-regression-tests.git - Unnamed repository; edit this file 'description' to name the repository.
index
:
github.com/marian-nmt/marian-regression-tests.git
line-by-line
master
master-updates
mjd/mergewithpmaster
mjd/types
optirun
pr-mjd-mbfstep
romang/new-metrics-finetune
usecpu
xml
Unnamed repository; edit this file 'description' to name the repository.
www-data
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
tests
/
training
Age
Commit message (
Expand
)
Author
2021-01-26
Merge branch 'master' into mjd/fp16.2
Roman Grundkiewicz
2021-01-26
Revert adagrad test to --clip-norm 1
Roman Grundkiewicz
2021-01-26
Add --clip-norm 1 to quant tests
Roman Grundkiewicz
2021-01-26
Revert quant tests to clip-norm 1
Roman Grundkiewicz
2021-01-25
Fix tests affected by changed logging formats
Roman Grundkiewicz
2021-01-25
Merge branch 'master' into mjd/fp16.2
Roman Grundkiewicz
2021-01-25
Merge branch 'master' into clip-norm-0
Roman Grundkiewicz
2021-01-25
Merge branch 'master' into mjd/fp16.2
Marcin Junczys-Dowmunt
2021-01-06
Revert/fix precision
Roman Grundkiewicz
2020-11-25
Update missing tests to clip-norm 0
Roman Grundkiewicz
2020-11-24
Add missing --clip-norm 0
Roman Grundkiewicz
2020-11-20
Revert "Update tests with quantization finetuning"
Roman Grundkiewicz
2020-11-16
Update tests with quantization finetuning
Roman Grundkiewicz
2020-11-16
Decrease precision
Roman Grundkiewicz
2020-11-13
Use clip-norm 0 in training/restarting
Roman Grundkiewicz
2020-11-13
Use clip-norm 0 in training/cost-functions
Roman Grundkiewicz
2020-11-13
Use clip-norm 0 in training/multi-gpu
Roman Grundkiewicz
2020-11-13
Use clip-norm 0 in training/basics
Roman Grundkiewicz
2020-11-13
Use clip-norm 0 in training/restoring/
Roman Grundkiewicz
2020-11-13
Use clip-norm 0 in training/features/
Roman Grundkiewicz
2020-11-12
Use clip-norm 0 and increase precisions for nematus
Roman Grundkiewicz
2020-11-12
Use clip-norm 0 in training/models/
Roman Grundkiewicz
2020-11-12
Use clip-norm 0 in training/validation/
Roman Grundkiewicz
2020-11-11
Merge pull request #69 from marian-nmt/quant-test
Roman Grundkiewicz
2020-11-11
test the resulting model, make sure it is quantized
Alham Fikri Aji
2020-11-11
add log-based quantization test
Alham Fikri Aji
2020-11-11
simplify quantized-model testing by using a shared train folder
Alham Fikri Aji
2020-11-10
Update expected outputs
pr-mjd-mbfstep
Roman Grundkiewicz
2020-11-10
Add tag #mini-batch-fit
Roman Grundkiewicz
2020-11-10
Update expected output and precision for optimized-opt
Roman Grundkiewicz
2020-11-10
Remove trailing whitespaces
Roman Grundkiewicz
2020-11-10
Add setup.sh and .gitignore
Roman Grundkiewicz
2020-11-10
Create train/ directory for vocabs
Roman Grundkiewicz
2020-11-10
Merge branch 'master' into quant-test
Roman Grundkiewicz
2020-11-10
correct the expected file
Alham Fikri Aji
2020-11-10
add regression tests for model quantization
Alham Fikri Aji
2020-11-10
add regression tests for model quantization
Alham Fikri Aji
2020-11-09
Add tests with --logical-epoch and --after
Roman Grundkiewicz
2020-10-28
Fix seed
Roman Grundkiewicz
2020-10-21
Update tests for restoring corpus and validation states
Roman Grundkiewicz
2020-10-21
Update tests for restarting optimizers
Roman Grundkiewicz
2020-10-21
Update tests for multi-gpu training and exp-smoothing
Roman Grundkiewicz
2020-10-21
Update tests for multi-gpu training
Roman Grundkiewicz
2020-10-21
Update tests for different models
Roman Grundkiewicz
2020-10-21
Update tests for other training features
Roman Grundkiewicz
2020-10-21
Update tests for data-weighting
Roman Grundkiewicz
2020-10-21
Update tests for cost-functions
Roman Grundkiewicz
2020-10-21
Update basic tests for --cost-type ce-sum
Roman Grundkiewicz
2020-09-02
Update test with toy vocab
Roman Grundkiewicz
2020-09-02
Fix misspelling
Roman Grundkiewicz
[next]