| .. |
|
backends
|
Opt model save and load (#126374)
|
2024-06-05 13:01:16 +00:00 |
|
repro
|
Opt model save and load (#126374)
|
2024-06-05 13:01:16 +00:00 |
|
variables
|
[dynamo][pippy] Hotfix for nn_module_stack for pippy usecase (#127972)
|
2024-06-05 20:14:50 +00:00 |
|
__init__.py
|
[easy][dynamo] Use disable_dynamo for torch.manual_seed (#126192)
|
2024-05-14 23:20:32 +00:00 |
|
_trace_wrapped_higher_order_op.py
|
|
|
|
bytecode_analysis.py
|
[dynamo] fix call_finally issue in Python 3.8 (#124122)
|
2024-04-16 08:36:20 +00:00 |
|
bytecode_transformation.py
|
[dynamo] utility to generate bytecode from template function (#127359)
|
2024-05-30 06:37:32 +00:00 |
|
cache_size.py
|
|
|
|
callback.py
|
[torch.compile] Provide capability to register callback on compile start/stop (#120764)
|
2024-02-29 07:37:52 +00:00 |
|
code_context.py
|
|
|
|
codegen.py
|
Implement native support for float inputs in Dynamo and ShapeEnv (#125325)
|
2024-05-14 04:10:01 +00:00 |
|
compiled_autograd.py
|
[compiled autograd] log in cpp using python logger (#126483)
|
2024-05-19 23:49:52 +00:00 |
|
comptime.py
|
Fix bug for comptime.get_local for cells/closures (#126637)
|
2024-05-20 17:51:28 +00:00 |
|
config.py
|
Properly detect nested torch function args (#127496)
|
2024-06-02 03:43:22 +00:00 |
|
convert_frame.py
|
[BE][Eazy] remove torch.torch.xxx usages (#127800)
|
2024-06-05 21:53:49 +00:00 |
|
create_parameter_op.py
|
functionalize storage resizing, minimal ppFSDP traceable forward (#122434)
|
2024-05-10 18:09:10 +00:00 |
|
current_scope_id.py
|
|
|
|
debug_utils.py
|
Remove removed ruff rule TRY200 (#126256)
|
2024-05-17 16:31:05 +00:00 |
|
decorators.py
|
Beef up the allow_in_graph docs (#127117)
|
2024-06-02 15:00:46 +00:00 |
|
device_interface.py
|
[ROCm] Triton upstream AMD backend integration (#121801)
|
2024-04-25 20:44:27 +00:00 |
|
eval_frame.py
|
Opt model save and load (#126374)
|
2024-06-05 13:01:16 +00:00 |
|
exc.py
|
[dynamo] Initial exception handling support (#126923)
|
2024-06-01 13:00:32 +00:00 |
|
external_utils.py
|
Revert "[BE] wrap deprecated function/class with typing_extensions.deprecated (#126898)"
|
2024-05-31 19:47:24 +00:00 |
|
funcname_cache.py
|
[dynamo] better file open method in funcname_cache (#125435)
|
2024-05-06 20:55:15 +00:00 |
|
guards.py
|
[dynamo][compile-time] Cache the child guard managers (#127377)
|
2024-05-31 04:23:56 +00:00 |
|
hooks.py
|
|
|
|
logging.py
|
|
|
|
mutation_guard.py
|
[dynamo] Unspec nn module when global backward hooks are present (#127802)
|
2024-06-04 18:25:46 +00:00 |
|
output_graph.py
|
Rewrite existing links to custom ops gdocs with the landing page (#127423)
|
2024-05-30 14:54:29 +00:00 |
|
polyfill.py
|
[Dynamo] Wrap set as SetVariable and support isdisjoint by polyfill (#126046)
|
2024-05-14 04:56:06 +00:00 |
|
profiler.py
|
|
|
|
replay_record.py
|
|
|
|
resume_execution.py
|
[Dynamo] make bytecode of resume function resemble natural bytecode (#126630)
|
2024-05-23 05:06:33 +00:00 |
|
side_effects.py
|
[dynamo] Bug fix for LOAD_GLOBAL and STORE_GLOBAL (#125002)
|
2024-04-28 15:24:17 +00:00 |
|
source.py
|
Teach FakeTensor to fill in item_memo when converting scalar CPU tensor (#126245)
|
2024-05-22 15:25:38 +00:00 |
|
symbolic_convert.py
|
[dynamo] Initial exception handling support (#126923)
|
2024-06-01 13:00:32 +00:00 |
|
tensor_version_op.py
|
Support torchbind op dispatch in python (#123367)
|
2024-04-19 17:17:27 +00:00 |
|
test_case.py
|
[dynamo, 3.12] enable dynamo on 3.12, enable most dynamo unittests on 3.12 (#123216)
|
2024-04-04 20:00:54 +00:00 |
|
test_minifier_common.py
|
|
|
|
testing.py
|
[dynamo] utility to generate bytecode from template function (#127359)
|
2024-05-30 06:37:32 +00:00 |
|
trace_rules.py
|
Revert "Add aten._unsafe_masked_index (#116491)"
|
2024-06-03 15:51:50 +00:00 |
|
types.py
|
[1/N][Easy] fix typo for usort config in pyproject.toml (kown -> known): sort stdlib (#127122)
|
2024-05-25 08:25:50 +00:00 |
|
utils.py
|
[dynamo] Unspec nn module when global backward hooks are present (#127802)
|
2024-06-04 18:25:46 +00:00 |