Training GPT-2 transformer language model with sentencepiece tokenizer

.. image:: https://img.shields.io/travis/lopuhin/transformer-lm/master.svg :target: https://travis-ci.org/lopuhin/transformer-lm :alt: Build Status

Training GPT-2 transformer language model on your own corpora with sentencepiece <https://github.com/google/sentencepiece>_ tokenization.

This repo contains a PyTorch implementation of GPT-2, which support multi-GPU training. It also contains a TensorFlow implementation in lm/gpt_2_tf, but it is not developed any more. They share the same data preparation scripts. TF training command is gpt-2-tf-train and needs TensorFlow 1.13. Documentation below is for PyTorch version.

.. contents::

Installation

Python 3.6+ is required with torch nightly or 1.6.0+. Working in a virtualenv is assumed below. Install <https://pytorch.org/get-started/locally/>__ appropriate version of pytorch first, and then::

pip install -r requirements.txt
python setup.py develop

Usage

Instructions are below. See also test/test_shakespeare.sh for a complete pipeline demo on a small corpus (takes a minute on a CPU).

Prepare data for training +++++++++++++++++++++++++

Corpus format: a directory with top-level train, valid and test folders. Each top-level folder may contain sub-folders. Inside them, there must be utf-8 encoded text files with .txt extension.

The commands to train sentencepiece model and encode the corpus support multiple corpora, in below examples we assume they can be listed as data/corpora-*.

1. Train sentencepiece model (sp-text.txt can be removed after running). This can consume a large amount of memory, adjust sentencepiece arguments as advised if needed (this is not supported in the sp-train command directly)::

   sp-train data/corpora-* sp-text.txt sp-model

2. Encode corpora, producing numpy files::

   sp-encode data/corpora-* sp-model.model data/encoded

Training ++++++++

Example command::

gpt-2 run-root data/encoded sp-model.model

run-root would contain model checkpoints and json-lines logs, which can be plotted in a jupyter notebook with json_log_plots.plot("run-root"), with number of tokens seen on the X axis.

Default hyperparameters correspond to released "small" GPT-2 model.

When multiple GPUs are available, they would be used for training with the help of torch.distributed.

If the path exists and --clean key is NOT passed, training would be resumed. Note that all parameters still need to be specified and model parameters need to match.

Notes on training parameters:

• --batch-size is per-GPU, so you don't need to re-tune it when changing number of GPUs, just use max that fits into memory.
• --g-accum-gradients is the global number of gradient accumulations, it must be divisible by the number of GPUs. Effective global batch size is always batch_size * g_accum_gradients.
• --lr does not need to be changed when changing --batch-size or --g-accum-gradients or number of GPUs or --n-ctx: loss is already scaled appropriately.

Inference +++++++++

Example command::

gpt-2-gen run-root "Artificial intelligence"

run-root would contain model checkpoints "Artificial intelligence" is the text prefix used as a starting point for generating tokens

Notes on inference parameters:

• --tokens-to-generate: number of tokens to generate, default is 42
• --top-k: number of token candidates to generate for each position (beam width), default is 8.

License & credits

License is MIT.

TensorFlow GPT-2 model is taken from https://github.com/openai/gpt-2/blob/master/src/model.py and TensorFlow GPT-2 training code is based on https://github.com/nshepperd/gpt-2/blob/finetuning/train.py

PyTorch port is based on original OpenAI code.

Test Shakespeare corpus under tests/shakespeare is from http://shakespeare.mit.edu under public domain.

See also OpenAI GPT-2 paper <https://d4mucfpksywv.cloudfront.net/better-language-models/language-models.pdf>_ and blog <https://openai.com/blog/better-language-models/>_.