diff options
author | Marcin Junczys-Dowmunt <junczys@amu.edu.pl> | 2017-11-26 18:27:17 +0300 |
---|---|---|
committer | GitHub <noreply@github.com> | 2017-11-26 18:27:17 +0300 |
commit | 050843dd917043bb5588ea431d02dcab68b567e6 (patch) | |
tree | 90465419c485fb2df78f8bd466598f363659f0dd | |
parent | 8762799683a3a7b9f564427e923297c499ab5486 (diff) |
Update README.md
-rw-r--r-- | training-basics/README.md | 38 |
1 files changed, 19 insertions, 19 deletions
diff --git a/training-basics/README.md b/training-basics/README.md index cb96331..8f4eb23 100644 --- a/training-basics/README.md +++ b/training-basics/README.md @@ -28,32 +28,32 @@ Next it executes a training run with `marian`: ``` ../../build/marian \ - --devices $GPUS \ - --type amun \ - --model model/model.npz \ - --train-sets data/corpus.bpe.ro data/corpus.bpe.en \ - --vocabs model/vocab.ro.yml model/vocab.en.yml \ - --dim-vocabs 66000 50000 \ - --mini-batch-fit -w 3000 \ - --layer-normalization --dropout-rnn 0.2 --dropout-src 0.1 --dropout-trg 0.1 \ - --early-stopping 5 \ - --valid-freq 10000 --save-freq 10000 --disp-freq 1000 \ - --valid-metrics cross-entropy translation \ - --valid-sets data/newsdev2016.bpe.ro data/newsdev2016.bpe.en \ - --valid-script-path ./scripts/validate.sh \ - --log model/train.log --valid-log model/valid.log \ - --seed 1111 --exponential-smoothing \ - --normalize=0.6 --beam-size 6 + --devices $GPUS \ + --type amun \ + --model model/model.npz \ + --train-sets data/corpus.bpe.ro data/corpus.bpe.en \ + --vocabs model/vocab.ro.yml model/vocab.en.yml \ + --dim-vocabs 66000 50000 \ + --mini-batch-fit -w 3000 \ + --layer-normalization --dropout-rnn 0.2 --dropout-src 0.1 --dropout-trg 0.1 \ + --early-stopping 5 \ + --valid-freq 10000 --save-freq 10000 --disp-freq 1000 \ + --valid-metrics cross-entropy translation \ + --valid-sets data/newsdev2016.bpe.ro data/newsdev2016.bpe.en \ + --valid-script-path ./scripts/validate.sh \ + --log model/train.log --valid-log model/valid.log \ + --overwrite --keep-best \ + --seed 1111 --exponential-smoothing \ + --normalize=1 --beam-size=12 --quiet-translation ``` After training (the training should stop if cross-entropy on the validation set -stops improving) a final model `model/model.avg.npz` is created from the 4 best -models on the validation sets (by element-wise averaging). This model is used +stops improving) the model with the highest translation validation score is used to translate the WMT2016 dev set and test set with `marian-decoder`: ``` cat data/newsdev2016.bpe.ro \ - | ../../build/marian-decoder -c model/model.npz.decoder.yml -m model/model.avg.npz -d $GPUS -b 12 -n \ + | ../../build/marian-decoder -c model/model.npz.best-translation.npz.decoder.yml -d $GPUS -b 12 -n \ | sed 's/\@\@ //g' \ | ../tools/moses-scripts/scripts/recaser/detruecase.perl \ | ../tools/moses-scripts/scripts/tokenizer/detokenizer.perl -l en \ |