..
backends
Remove unused Python variables in torch/[_-a]* ( #133492 )
2024-12-12 17:39:14 +00:00
polyfills
[dynamo][pytree][3/N] make CXX pytree traceable: tree_map / tree_map_ ( #137399 )
2024-12-12 18:05:25 +00:00
repro
Remove unused Python variables in torch/[_-a]* ( #133492 )
2024-12-12 17:39:14 +00:00
variables
Prevent users from seeing hardcoded print stmt when hypothesis is not installed ( #142398 )
2024-12-17 16:59:05 +00:00
__init__.py
[ca] add graph id ( #141906 )
2024-12-14 03:02:06 +00:00
_trace_wrapped_higher_order_op.py
bytecode_analysis.py
bytecode_transformation.py
cache_size.py
callback.py
code_context.py
codegen.py
compiled_autograd.py
[ca] add graph id ( #141906 )
2024-12-14 03:02:06 +00:00
comptime.py
Remove unused Python variables in torch/[_-a]* ( #133492 )
2024-12-12 17:39:14 +00:00
config.py
Support garbage collection after pt2 compilation ( #143364 )
2024-12-18 07:25:11 +00:00
convert_frame.py
Support garbage collection after pt2 compilation ( #143364 )
2024-12-18 07:25:11 +00:00
create_parameter_op.py
current_scope_id.py
debug_utils.py
Remove unused Python variables in torch/[_-a]* ( #133492 )
2024-12-12 17:39:14 +00:00
decorators.py
[Dynamo] only import einops if version is lower than 0.7.0 ( #142847 )
2024-12-17 20:50:25 +00:00
device_interface.py
distributed.py
eval_frame.py
[dynamo] disable eval frame callback around most of _TorchDynamoContext wrapper function ( #143211 )
2024-12-16 18:38:58 +00:00
exc.py
Rename 'cache limit' to 'recompile limit' ( #141542 )
2024-12-11 05:05:11 +00:00
external_utils.py
Deprecate torch._utils.is_compiling() ( #127690 )
2024-12-08 22:55:36 +00:00
funcname_cache.py
graph_deduplication.py
[Dynamo] Initial deduplication pass impl ( #141383 )
2024-12-11 02:22:21 +00:00
graph_region_tracker.py
[Dynamo] add debug logging for graph region expansion ( #141382 )
2024-12-11 02:22:21 +00:00
guards.py
[dynamo] implement framelocals mapping as c++ object ( #140063 )
2024-12-17 18:54:27 +00:00
hooks.py
logging.py
metrics_context.py
[logging] Log runtime autotuning timing to scuba ( #141919 )
2024-12-13 21:22:13 +00:00
mutation_guard.py
output_graph.py
refactor tensorify restart logic to use sources ( #141517 )
2024-12-11 07:15:39 +00:00
pgo.py
filelock: Make waitcounter variant to use ( #139816 )
2024-12-12 01:18:34 +00:00
profiler.py
replay_record.py
resume_execution.py
[dynamo] implement framelocals mapping as c++ object ( #140063 )
2024-12-17 18:54:27 +00:00
side_effects.py
source.py
symbolic_convert.py
dynamo tracing perf: cache on import_source: 52.9 -> 52.58 ( #143058 )
2024-12-13 18:20:48 +00:00
tensor_version_op.py
test_case.py
test_minifier_common.py
Remove unused Python variables in torch/[_-a]* ( #133492 )
2024-12-12 17:39:14 +00:00
testing.py
[dynamo] implement framelocals mapping as c++ object ( #140063 )
2024-12-17 18:54:27 +00:00
trace_rules.py
Revert "[ROCm] CK Flash Attention Backend ( #138947 )"
2024-12-17 16:46:57 +00:00
types.py
utils.py
Support garbage collection after pt2 compilation ( #143364 )
2024-12-18 07:25:11 +00:00