geowatch.cli.experimental.recommend_size_adjustments module¶

class geowatch.cli.experimental.recommend_size_adjustments.RecommendSizeAdjustmentsCLI(*args, **kwargs)[source]¶

Bases: DataConfig

Helper to recommend adjustments to network size parameters

Example

MAX_STEPS=10000 MAX_EPOCHS=720 TRAIN_BATCHES_PER_EPOCH=2666 ACCUMULATE_GRAD_BATCHES=32 BATCH_SIZE=6 TRAIN_ITEMS_PER_EPOCH=$(python -c “print($TRAIN_BATCHES_PER_EPOCH * $BATCH_SIZE)”)

python -m geowatch.cli.experimental.recommend_size_adjustments –MAX_STEPS=$MAX_STEPS –MAX_EPOCHS=$MAX_EPOCHS –BATCH_SIZE=$BATCH_SIZE –ACCUMULATE_GRAD_BATCHES=$ACCUMULATE_GRAD_BATCHES –TRAIN_BATCHES_PER_EPOCH=”$TRAIN_BATCHES_PER_EPOCH” –TRAIN_ITEMS_PER_EPOCH=”$TRAIN_ITEMS_PER_EPOCH”

Valid options: []

Parameters:

*args – positional arguments for this data config
**kwargs – keyword arguments for this data config

classmethod main(cmdline=1, **kwargs)[source]¶

Example

>>> # xdoctest: +SKIP
>>> from geowatch.cli.experimental.recommend_size_adjustments import *  # NOQA
>>> cmdline = 0
>>> kwargs = dict()
>>> cls = RecommendSizeAdjustmentsCLI
>>> cls.main(cmdline=cmdline, **kwargs)

default = {'ACCUMULATE_GRAD_BATCHES': <Value(1)>, 'BATCH_SIZE': <Value(None)>, 'MAX_EPOCHS': <Value(None)>, 'MAX_STEPS': <Value(None)>, 'TRAIN_BATCHES_PER_EPOCH': <Value(None)>, 'TRAIN_ITEMS_PER_EPOCH': <Value(None)>}¶

geowatch.cli.experimental.recommend_size_adjustments.main(cmdline=1, **kwargs)¶

Example

>>> # xdoctest: +SKIP
>>> from geowatch.cli.experimental.recommend_size_adjustments import *  # NOQA
>>> cmdline = 0
>>> kwargs = dict()
>>> cls = RecommendSizeAdjustmentsCLI
>>> cls.main(cmdline=cmdline, **kwargs)