Open shresthpaul133 opened 6 years ago
@oahziur Sir please help.
To train the vi to en training set follow the following steps
pip install -U tf-nightly
git clone https://github.com/tensorflow/nmt/
cd nmt
nmt/scripts/download_iwslt15.sh /tmp/nmt_data
mkdir /tmp/nmt_model
python -m nmt.nmt \ --src=vi --tgt=en \ --vocab_prefix=/tmp/nmt_data/vocab \ --train_prefix=/tmp/nmt_data/train \ --dev_prefix=/tmp/nmt_data/tst2012 \ --test_prefix=/tmp/nmt_data/tst2013 \ --out_dir=/tmp/nmt_model \ --num_train_steps=12000 \ --steps_per_stats=100 \ --num_layers=2 \ --num_units=128 \ --dropout=0.2 \ --metrics=bleu
@oahziur Sir any idea on how to tackle this problem ? I corrected the data and trained it on my own data set.
I also ran into similar issues and I resolved it making these changes https://github.com/tensorflow/nmt/pull/401
I need some help in running the program, I followed all the steps and trained this to 1000 steps. Now when I am running this so that I can get the required output, and can check the results. I am just getting the output as FAILED (failures = 3). So how to resolve this. P.S. I'm using the following command.
python -m nmt.model_test