Also add the --batchsize option to training apps

stickeritis / sticker

Succeeded by SyntaxDot: https://github.com/tensordot/syntaxdot

Other

25 stars 2 forks source link

Closed danieldk closed 4 years ago

danieldk commented 4 years ago

The pipeline apps were already using the --batchsize option in place of the sticker configuration file. Make things consistent again by doing the same in the train apps.
Add a warning that the 'model.batch_size' option is not used anymore if it is present in the configuration.

Fixes #151.

danieldk commented 4 years ago

Still have to test this ;).