aibedo.train.run_model

aibedo.train.run_model(config: omegaconf.dictconfig.DictConfig) → float

This function runs/trains/tests the model.

Note

It is recommended to call this function by running its underlying script, aibedo.train.py, as this will enable you to make the best use of the command line integration with Hydra. For example, you can easily train an MLP for 10 epochs on the CPU with:

>>>  python train.py trainer.max_epochs=10 trainer.gpus=0 model=mlp logger=none callbacks=default

Parameters

config – A DictConfig object generated by hydra containing the model, data, callbacks & trainer configuration.

Returns

float – the best model score reached while training the model. E.g. “val/mse”, the mean squared error on the validation set.