KeyError: 'logger' when executing a NER trainer

[ines (Ines Montani)]

Hi! It looks like this error happens if the base config doesn't contain a logger – we already have a fix and will be releasing v1.11.1 shortly, but in the meantime, you can just comment out this section in train.py (you can run prodigy stats to find the location of your Prodigy installation).

See this comment for more background:

1.11.0: Incorrect generation of config files?

Hi Tom, Sorry for not having been entirely clear in my previous reply. We are aware of the fact that the functionality changed since the nightly version, as we've specifically implemented it so that parts of the config of the base_model would be copied over, including the training parameters. This is to better support transformer-based pipelines, which need different training parameters. In general, if you specify a base_model then Prodigy will try to copy as much as possible from that model in…