aitune.torch.jit.config
Configuration for JIT module.
Module Contents
Classes
Functions
Data
API
Dataclass
Configuration for JIT module.
batch_axis_required
cache_dir
detect_graph_breaks
device
dry_run
dry_run_failure_probability
max_depth_level
min_parameters
min_samples
mode
patch_exclude
skip_modules
strategy
Post init.
Reset all options to their default values (e.g. for test isolation).
Return the tune strategy to use for JIT tuning.
When strategy is set explicitly it is returned as-is. Otherwise the dynamic
resolver selects JIT backends from the module’s execution properties and builds a
MaxThroughputStrategy.
Parameters:
module
Module that will be tuned.
Bases: enum.Enum
Mode for JIT execution.
INSPECT
TUNE_DEFERRED
TUNE_EAGER
Return the default strategy.