the-algorithm/twml
2023-04-11 09:04:51 -04:00
..
libtwml Optimization 2023-04-11 09:04:51 -04:00
twml Optimization 2023-04-11 09:04:51 -04:00
twml_common Optimization 2023-04-11 09:04:51 -04:00
BUILD Optimization 2023-04-11 09:04:51 -04:00
README.md Optimization 2023-04-11 09:04:51 -04:00
setup.cfg Optimization 2023-04-11 09:04:51 -04:00
setup.py Optimization 2023-04-11 09:04:51 -04:00

TWML


Note: twml is no longer under development. Much of the code here is not out of date and unused. It is included here for completeness, because twml is still used to train the light ranker models (see src/python/twitter/deepbird/projects/timelines/scripts/models/earlybird/README.md)

TWML is one of Twitter's machine learning frameworks, which uses Tensorflow under the hood. While it is mostly deprecated, it is still currently used to train the Earlybird light ranking models ( see src/python/twitter/deepbird/projects/timelines/scripts/models/earlybird/train.py). The most relevant part of this is the DataRecordTrainer class, which is where the core training logic resides.