The code is modified from repo.

Major improvement:

• support fix embedding or finetune embedding
• fine tune part of embeddings in hypothese
• Antomatically inference some parameter from the checkpoint, deprecated the test config.

Preprocess the data

Before the downloaded corpus and embeddings can be used in the ESIM model, they need to be preprocessed. This can be done with the preprocess_data.py script in the scripts/ folder of this repository.

The script's usage is the following:

preprocess_data.py [-h] [--config CONFIG]

where config is the path to a configuration file defining the parameters to be used for preprocessing. A default configuration file can be found in the config/ folder of this repository.

Train the model

The train_model.py script can be used to train the ESIM model on some training data and validate it on some validation data.

The script's usage is:

train_model.py [-h] [--config CONFIG] [--checkpoint CHECKPOINT]

where config is a configuration file (a default one is located in the config/ folder), and checkpoint is an optional checkpoint from which training can be resumed. Checkpoints are created by the script after each training epoch, with the name esim_*.pth.tar, where '*' indicates the epoch's number.

Test the model

The test_model.py script can be used to test the model on some test data.

Its usage is:

test_model.py [-h]