v1.2.1

Minor changes:

• Fix a bug that could slightly hurt performance on WIC and WSC (#916).

v1.2.0

Highlighted changes:

• Add support for RoBERTa, XLM, and GPT-2 via pytorch_transformers 1.2.
• Add support for pip installation (and moved the body of main.py and the config directory to accomodate that change).
• Fix a bug that produced invalid micro/macro average scores during validation.

Minor changes:

• Refactor old GPT (v1) implementation to use pytorch_transformers.
• Make the code that adds git status information to logs more robust.
• Minor cleanup to data loading and to MNLI data handling logic.
• Fix a short-lived bug invalidating hypothesis-only MNLI results.
• Restore (partial) support for sequence-to-sequence tasks, though with no fully supported demonstration tasks in place yet.

Dependency changes:

• Updated requirement pytorch_transformers to 1.2.0.
• Updated requirement to NLTK 3.4.5 to avoid a potential security issue.

v1.1.0

We expect another release within a week or two that will add support for RoBERTa (see #890), but this is a quick intermediate release now that XLNet support is stable/working.

Highlighted changes:

• Full support for XLNet and the whole-word-masking variants of BERT.
• Many small improvements to Google Cloud Platform/Kubernetes/Docker support.
• Add small but handy option to automatically delete checkpoints when a job finishes.
• max_vals is now used when computing warmup time with optimizers that use warmup.
• New auto option for tokenizer chooses an appropriate tokenizer for any given input module.
• Some internal changes to how <SOS>/<EOS>/[SEP]/[CLS] tokens are handled during task preprocessing. This will require small changes to custom task code along the lines of what is seen in #845.

Dependency changes:

• AllenNLP 0.8.4 now required
• pytorch_transformers 1.0 now required when using BERT or XLNet.

Warnings:

• Upgrading to 1.1 will break existing checkpoints for BERT-based models.

v1.0.1

Bug fixes:

• Addresses an issue that prevented temporary checkpoints from being deleted.

v1.0.0

The first stable release of jiant.

Highlighted changes:

• Support for the SuperGLUE v2.0 set of tasks, including all the baselines discussed in the SuperGLUE paper.
• A simpler and more standard code structure.
• Cleaner, more-readable logs.
• Simplified logic for checkpointing and evaluation, with fewer differences between pretraining and target task training.
• Fewer deprecated/unsupported modules.
• Many small bug fixes and improvements to errors and warnings.

Dependency changes:

• Upgrade to AllenNLP 0.8.4, which adds the option to use the GitHub development version of pytorch-pretrained-bert, and with it, the whole-word-masking variants of BERT.

Warnings:

• Upgrading from 0.9 to 1.0 will break most older model checkpoints and cached preprocessed data.

"Can You Tell Me How to Get Past Sesame Street?" code

This release contains code to recreate part of the experiments from the paper "Can You Tell Me How to Get Past Sesame Street? Sentence-Level Pretraining Beyond Language Modeling". For the remaining experiments, see this branch.

v0.9.1

A release with fixes to SuperGLUE baseline scripts/configurations.

v0.9.0

The initial work-in-progress release coinciding with the launch of SuperGLUE.

Highlights:

We currently support two-phase training (pretraining and target task training) using various shared encoders, including:

• BERT
• OpenAI GPT
• Plain Transformer
• Ordered Neurons (ON-LSTM) Grammar Induction Model
• PRPN Grammar Induction Model

We also have support for SuperGLUE baselines, sentence encoder probing experiments, and STILTS-style training.

Examples

They can be found in https://github.com/nyu-mll/jiant/tree/master/config/examples

"ELMo's Friends" paper experiment code

s2s decoder update (make more params active; add projection layer) (#…

…384)

* quick fix for mt

* [veryminor] add mt_attention parameter to defaults.conf

* readme: mt->s2s

* [s2sdecoder] add bottleneck layer

* [s2sdecoder] make all parameters active

* clean beamsearch

* pull/384: address comments

* fix merge

* Add s2s_ prefix to opts; add projection documentation in code

* Rename s2s configs + refactor param dict construction; fix scheduled sampling bug; fix projected dim bug

* Redo decoder param construction

* Remove unneeded code

* Remove hasattr calls; fix bug when no attn

* Fix wrong param name