Configuration File¶
Besides the arguments provided to the command line tools, detailed training and
prediction configuration is specified in a yaml
file. The configuration file contains
two sections, dataset
and train
.
Dataset section¶
See the dataset loader configuration for more details.
Train section¶
The train
section defines the neural network training hyper-parameters, by specifying
subsections, model
, loss
, optimizer
, preprocess
and other training
hyper-parameter, including epochs
and save_period
. See an
example configuration,
with comments on the available options in each subsection.
This section is highly application-specific. More examples can be found in DeepReg Demos.