No saving of models and datasets

DeNederlandscheBank / nqm

A Transformer-based Machine for answering questions on insurance companies

MIT License

0 stars 0 forks source link

No saving of models and datasets #2

Closed jm-glowienke closed 3 years ago

jm-glowienke commented 3 years ago

Current pipeline (branch iss1) overwrites existing models and datasets (data/nqm/dictionairies). This is not ideal for performance control and reproducibility.

jm-glowienke commented 3 years ago

I added job-id generator to the pipeline. This can be read in to the python files and be used to give unique names to models and datasets. By using the same identifier, one can easily control what belongs together.