Skip to article frontmatterSkip to article content
Site not loading correctly?

This may be due to an incorrect BASE_URL configuration. See the MyST Documentation for reference.

The droulette package

The main use cases are documentet as the Machine Learning Pipeline. This page documents a few additional features.

Training of individual models

Make a config TOML file, specifying the model and the hyperparamers. An example is provided as Test/experiment01.toml

python -m droulette.model --config Test/experiment01.toml

Plot training development

To plot the development of the loss function over epochs, the following module can be used. In batch, it creates a plot file training.svg from the training_log.csv files in each directory.

python -m droulette.diagnostic --alldirs _test/experiment???

It can also be run on individual log files, without the --alldirs option.