Training

This section explains how to train ISLR models using the existing datasets and models.

Config-based training

For examples on how to use the datasets and models in configs, click here.

After you have a config ready, run the following python snippet:

import omegaconf
from openhands.apis.classification_model import ClassificationModel
from openhands.core.exp_utils import get_trainer

cfg = omegaconf.OmegaConf.load("path/to/config.yaml")
trainer = get_trainer(cfg)

model = ClassificationModel(cfg=cfg, trainer=trainer)
model.init_from_checkpoint_if_available()
model.fit()

• This will automatically do all the setup, and start the training for you!

• The best checkpoints will also be dumped based on validation from each epoch.

• Feel free to play with the different parameters in the existing configs