tensorflow
Is it possible to output top-k predictions?
If I provide one sentence, is it possible to inference top-k predictions, not only the best prediction?
Looks like this work provides this function, but I didn't success on it, and always get only the best prediction.
I put only one sentence in /tmp/inference/one.rct, and set --num_translations_per_input=5 in inference mode, but get only one prediction, not 5 candidates.
I use tensorflow-nightly on the master branch, commit 365e73.
Could you give me some directions?
training script
python -m nmt.nmt \
--attention=scaled_luong \
--src=vi --tgt=en \
--vocab_prefix=/tmp/data/token \
--train_prefix=/tmp/data/train \
--dev_prefix=/tmp/data/valid \
--test_prefix=/tmp/data/test \
--out_dir=/tmp/model-top-k \
--num_train_steps=10000 \
--steps_per_stats=100 \
--num_layers=2 \
--num_units=1024 \
--dropout=0.2 \
--metrics=bleu \
--optimizer=sgd \
--learning_rate=1.0 \
--start_decay_step=5000 \
--decay_steps=10 \
--encoder_type=bi \
--beam_width=10
inference script
python -m nmt.nmt \
--out_dir=/tmp/model-top-k \
--inference_input_file=/tmp/inference/one.rct \
--inference_output_file=/tmp/inference/one_pred.prd \
--num_translations_per_input=5 \
--beam_width=10
I did a dirty fix by adding:
hparams.num_translations_per_input = flags.num_translations_per_input
hparams.beam_width = flags.beam_width
in nmt.py, in run_main(.) just after hparams is loaded, it works just fine.
Thanks @ttrouill It also works from my site.
Looks like: inference uses hparams from training, and it is not updated for inference params.
https://github.com/tensorflow/nmt/blob/master/nmt/nmt.py#L577
Are you going to create a PR? I can create one if you are not available. Let me know your preference.
Go ahead :) Though it should probably be something cleaner and more general (to pass other unpassed parameters too) than what I proposed.
