Repository files navigation

Documentation

This is a framework for sequence-to-sequence (seq2seq) models implemented inPyTorch. The framework has modularized and extensible components for seq2seq models, training and inference, checkpoints, etc. This is an alpha release. We appreciate any kind of feedback or contribution.

• Compatible with PyTorch 0.4
• Added support for pre-trained word embedding

Seq2seq is a fast evolving field with new techniques and architectures being published frequently. The goal of this library is facilitating the development of such techniques and applications. While constantly improving the quality of code and documentation, we will focus on the following items:

• Evaluation with benchmarks such as WMT machine translation, COCO image captioning, conversational models, etc;
• Provide more flexible model options, improving the usability of the library;
• Adding latest architectures such as the CNN based model proposed byConvolutional Sequence to Sequence Learning and the transformer model proposed byAttention Is All You Need;
• Support features in the new versions of PyTorch.

This package requires Python 2.7 or 3.6. We recommend creating a new virtual environment for this project (using virtualenv or conda).

• Numpy:pip install numpy (Referhere for problem installing Numpy).
• PyTorch: Refer toPyTorch website to install the version w.r.t. your environment.

Currently we only support installation from source code using setuptools. Checkout the source code and run the following commands:

pip install -r requirements.txtpython setup.py install

If you already had a version of PyTorch installed on your system, please verify that the active torch package is at least version 0.1.11.

# Run script to generate the reverse toy dataset# The generated data is stored in data/toy_reverse by defaultscripts/toy.sh

TRAIN_PATH=data/toy_reverse/train/data.txtDEV_PATH=data/toy_reverse/dev/data.txt# Start trainingpython examples/sample.py --train_path $TRAIN_PATH --dev_path $DEV_PATH

It will take about 3 minutes to train on CPU and less than 1 minute with a Tesla K80. Once training is complete, you will be prompted to enter a new sequence to translate and the model will print out its prediction (use ctrl-C to terminate). Try the example below!

Input:  1 3 5 7 9Expected output: 9 7 5 3 1 EOS

Checkpoints are organized by experiments and timestamps as shown in the following file structure

experiment_dir+-- input_vocab+-- output_vocab+-- checkpoints|  +-- YYYY_mm_dd_HH_MM_SS   |  +-- decoder   |  +-- encoder   |  +-- model_checkpoint

The sample script by default saves checkpoints in theexperiment folder of the root directory. Look at the usages of the sample code for more options, including resuming and loading from checkpoints.

• WMT Machine Translation (Coming soon)

If you have any questions, bug reports, and feature requests, pleaseopen an issue on Github. For live discussions, please go to ourGitter lobby.

We appreciate any kind of feedback or contribution. Feel free to proceed with small issues like bug fixes, documentation improvement. For major contributions and new features, please discuss with the collaborators in corresponding issues.

We are using 4-week release cycles, where during each cycle changes will be pushed to thedevelop branch and finally merge to themaster branch at the end of each cycle.

We setup the development environment usingVagrant. Runvagrant up with our 'Vagrantfile' to get started.

The following tools are needed and installed in the development environment by default:

• Git
• Python
• Python packages: nose, mock, coverage, flake8

The quality and the maintainability of the project is ensured by comprehensive tests. We encourage writing unit tests and integration tests when contributing new codes.

Locally please runnosetests in the package root directory to run unit tests. We use TravisCI to require that a pull request has to pass all unit tests to be eligible to merge. Seetravis configuration for more information.

We followPEP8 for code style. Especially the style of docstrings is important to generate documentation.

• Local: Run the following commands in the package root directory

# Python syntax errors or undefined namesflake8 . --count --select=E901,E999,F821,F822,F823 --show-source --statistics# Style checksflake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics

• Github: We useCodacy to check styles on pull requests and branches.