The Natural Language Decathlon is a multitask challenge that spans ten tasks: question answering, machine translation, summarization, natural language inference, sentiment analysis, semantic role labeling, zero-shot relation extraction, goal-oriented dialogue, semantic parsing, and commonsense pronoun resolution. Each task is cast as question answering, which makes it possible to use our new Multitask Question Answering Network (MQAN). This model jointly learns all tasks in decaNLP without any task-specific modules or parameters in the multitask setting. For a more thorough introduction to decaNLP, see the main website, our blog post, or the paper.

Leaderboard

Model	decaNLP	SQuAD	IWSLT	CNN/DM	MNLI	SST	QA‑SRL	QA‑ZRE	WOZ	WikiSQL	MWSC
MQAN	590.5	74.4	18.6	24.3	71.5	87.4	78.4	37.6	84.8	64.8	48.7
S2S	513.6	47.5	14.2	25.7	60.9	85.9	68.7	28.5	84.0	45.8	52.4

Getting Started

First, make sure you have docker and nvidia-docker installed. Then build the docker image:

cd dockerfiles && docker build -t decanlp . && cd -

You will also need to make a .data directory and move the examples for the Winograd Schemas into it:

mkdir .data/schema
mv local_data/schema.txt .data/schema/

You can run a command inside the docker image using

nvidia-docker run -it --rm -v `pwd`:/decaNLP/ -u $(id -u):$(id -g) decanlp bash -c "COMMAND"

Training

For example, to train a Multitask Question Answering Network (MQAN) on the Stanford Question Answering Dataset (SQuAD):

nvidia-docker run -it --rm -v `pwd`:/decaNLP/ -u $(id -u):$(id -g) decanlp bash -c "python /decaNLP/train.py --train_tasks squad --gpu DEVICE_ID"

To multitask with the fully joint, round-robin training described in the paper, you can add multiple tasks:

nvidia-docker run -it --rm -v `pwd`:/decaNLP/ -u $(id -u):$(id -g) decanlp bash -c "python /decaNLP/train.py --train_tasks squad iwslt.en.de --train_iterations 1 --gpu DEVICE_ID"

To train on the entire Natural Language Decathlon:

nvidia-docker run -it --rm -v `pwd`:/decaNLP/ -u $(id -u):$(id -g) decanlp bash -c "python /decaNLP/train.py --train_tasks squad iwslt.en.de cnn_dailymail multinli.in.out sst srl zre woz.en wikisql schema --train_iterations 1 --gpu DEVICE_ID"

Tensorboard

If you would like to make use of tensorboard, run (typically in a tmux pane or equivalent):

docker run -it --rm -p 0.0.0.0:6006:6006 -v `pwd`:/decaNLP/ decanlp bash -c "tensorboard --logdir /decaNLP/results"

If you are running the server on a remote machine, you can run the following on your local machine to forward to http://localhost:6006/:

ssh -4 -N -f -L 6006:127.0.0.1:6006 YOUR_REMOTE_IP

If you are having trouble with the specified port on either machine, run lsof -if:6006 and kill the process if it is unnecessary. Otherwise, try changing the port numbers in the commands above. The first port number is the port the local machine tries to bind to, and and the second port is the one exposed by the remote machine (or docker container).

Caveats During Training

• On a single NVIDIA Volta GPU, the code should take about 3 days to complete 500k iterations. These should be sufficient to approximately reproduce the experiments in the paper. Training for about 7 days should be enough to fully replicate those scores.
• The model can be resumed using stored checkpoints using --load <PATH_TO_CHECKPOINT> and --resume. By default, models are stored every --save_every iterations in the results/ folder tree.
• During training, validation can be slow! Especially when computing ROUGE scores. Use the --val_every flag to change the frequency of validation.
• If you run out of memory, reduce --train_batch_tokens and --val_batch_size.
• The first time you run, the code will download and cache all considered datasets. Please be advised that this might take a while, especially for some of the larger datasets.

Evaluation

You can evaluate a model for a specific task with EVALUATION_TYPE as validation or test:

nvidia-docker run -it --rm -v `pwd`:/decaNLP/ -u $(id -u):$(id -g) decanlp bash -c "python /decaNLP/predict.py --evaluate EVALUATION_TYPE --path PATH_TO_CHECKPOINT_DIRECTORY --gpu DEVICE_ID --tasks squad"

or evaluate on the entire decathlon by removing any task specification:

nvidia-docker run -it --rm -v `pwd`:/decaNLP/ -u $(id -u):$(id -g) decanlp bash -c "python /decaNLP/predict.py --evaluate EVALUATION_TYPE --path PATH_TO_CHECKPOINT_DIRECTORY --gpu DEVICE_ID"

For test performance, please use the original SQuAD, MultiNLI, and WikiSQL evaluation systems.

Pretrained Models

This model is the best MQAN trained on decaNLP so far. It was trained first on SQuAD and then on all of decaNLP. You can obtain this model and run it on the validation sets with the following.

wget https://s3.amazonaws.com/research.metamind.io/decaNLP/pretrained/mqan_decanlp_qa_first.tar.gz
tar -xvzf mqan_decanlp_qa_first.tar.gz
nvidia-docker run -it --rm -v `pwd`:/decaNLP/  decanlp bash -c "python /decaNLP/predict.py --evaluate validation --path /decaNLP/mqan_decanlp_qa_first --checkpoint_name model.pth --gpu 0"

This model is the best MQAN trained on WikiSQL alone. It surpassed the previous state-of-the-art performance by several points on that task.

wget https://s3.amazonaws.com/research.metamind.io/decaNLP/pretrained/mqan_wikisql.tar.gz
tar -xvzf mqan_wikisql.tar.gz
nvidia-docker run -it --rm -v `pwd`:/decaNLP/  decanlp bash -c "python /decaNLP/predict.py --evaluate validation --path /decaNLP/mqan_wikisql --checkpoint_name model.pth --gpu 0 --tasks wikisql"
nvidia-docker run -it --rm -v `pwd`:/decaNLP/  decanlp bash -c "python /decaNLP/predict.py --evaluate test --path /decaNLP/mqan_wikisql --checkpoint_name model.pth --gpu 0 --tasks wikisql"
docker run -it --rm -v `pwd`:/decaNLP/  decanlp bash -c "python /decaNLP/convert_to_logical_forms.py /decaNLP/.data/ /decaNLP/mqan_wikisql/model/validation/wikisql.txt /decaNLP/mqan_wikisql/model/validation/wikisql.ids.txt /decaNLP/mqan_wikisql/model/validation/wikisql_logical_forms.jsonl valid"
docker run -it --rm -v `pwd`:/decaNLP/  decanlp bash -c "python /decaNLP/convert_to_logical_forms.py /decaNLP/.data/ /decaNLP/mqan_wikisql/model/test/wikisql.txt /decaNLP/mqan_wikisql/model/test/wikisql.ids.txt /decaNLP/mqan_wikisql/model/test/wikisql_logical_forms.jsonl test"
git clone https://github.com/salesforce/WikiSQL.git #[email protected]:salesforce/WikiSQL.git for ssh
cd WikiSQL
git checkout decanlp_single_model  # necessary until https://github.com/salesforce/WikiSQL/pull/23 is merged
cd ..
docker run -it --rm -v `pwd`:/decaNLP/  decanlp bash -c "python /decaNLP/WikiSQL/evaluate.py /decaNLP/.data/wikisql/data/dev.jsonl /decaNLP/.data/wikisql/data/dev.db /decaNLP/mqan_wikisql/model/validation/wikisql_logical_forms.jsonl" # assumes that you have data stored in .data
docker run -it --rm -v `pwd`:/decaNLP/  decanlp bash -c "python /decaNLP/WikiSQL/evaluate.py /decaNLP/.data/wikisql/data/dev.jsonl /decaNLP/.data/wikisql/data/dev.db /decaNLP/mqan_wikisql/model/test/wikisql_logical_forms.jsonl" # assumes that you have data stored in .data

Inference on a Custom Dataset

Using a pretrained model or a model you have trained yourself, you can run on new, custom datasets easily by following the instructions below. In this example, we use the checkpoint for the best MQAN trained on the entirety of decaNLP (see the section on Pretrained Models to see how to get this checkpoint) to run on my_custom_dataset.

mkdir .data/my_custom_dataset/
touch .datda/my_custom_dataset/val.jsonl
#TODO add examples line by line to val.jsonl in the form of a JSON dict: {"context": "The answer is answer.", "question": "What is the answer?", "answer": "answer"}
nvidia-docker run -it --rm -v `pwd`:/decaNLP/  decanlp bash -c "python /decaNLP/predict.py --evaluate valid --path /decaNLP/mqan_decanlp_qa_first --checkpoint_name model.pth --gpu 0 --tasks my_custom_dataset"

Citation

If you use this in your work, please cite The Natural Language Decathlon: Multitask Learning as Question Answering.

@article{McCann2018decaNLP,
  title={The Natural Language Decathlon: Multitask Learning as Question Answering},
  author={Bryan McCann and Nitish Shirish Keskar and Caiming Xiong and Richard Socher},
  journal={arXiv preprint arXiv:1806.08730},
  year={2018}
}

Contact

Contact: [email protected] and [email protected]