MCPcopy Create free account
hub / github.com/VincentHancoder/SSGD / auto_scale_lr

Function auto_scale_lr

mmdet/apis/train.py:71–114  ·  view source on GitHub ↗

Automatically scaling LR according to GPU number and sample per GPU. Args: cfg (config): Training config. distributed (bool): Using distributed or not. logger (logging.Logger): Logger.

(cfg, distributed, logger)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers 1

train_detectorFunction · 0.85

Calls 2

getMethod · 0.80
infoMethod · 0.45

Tested by

no test coverage detected