naivenmt

An experimental implementation of NMT using tensorflow.

Actually, this project is a refactor of tensorflow/nmt, with some improvements:

• Better code organization
• Export model with estimator API
• Transformer as described in Attention is all you need

See more about the GNMT's architecture and details, to tensorflow/nmt's README

All kinds of contributions are welcome.

• Usage
  • Train
  • Evaluate
  • Predict
  • Export

Usage

Almost all the params are the same as tensorflow/nmt except --mode. It's easy to use for those who have an experience for tensroflow/nmt.

For new users, the usage is simple, too. Here are some examples.

Train

python3 -m naivenmt.naivenmt \
    --mode=train \
    --src=en --tgt=vi \
    --out_dir=/tmp/model \
    --train_prefix=$YOUR_DATA_DIR/train \
    --dev_prefix=$YOUR_DATA_DIR/dev \
    --test_prefix=$YOUT_DATA_DIR/test \
    --vocab_prefix=$YOUR_DATA_DIR/vocab \
    --num_train_steps=10000 \
    ...(other params)

See the default values for hparams, to arguments.py

Evaluate

Predict

Export

License

 Copyright 2018 luozhouyang

 Licensed under the Apache License, Version 2.0 (the "License");
 you may not use this file except in compliance with the License.
 You may obtain a copy of the License at

     http://www.apache.org/licenses/LICENSE-2.0

 Unless required by applicable law or agreed to in writing, software
 distributed under the License is distributed on an "AS IS" BASIS,
 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 See the License for the specific language governing permissions and
 limitations under the License.