pytorch/torch/_dynamo
2024-10-09 20:08:01 +00:00
..
backends [Dynamo] Ensure torch function modes are dispatched on builtin ops (#137117) 2024-10-09 02:29:40 +00:00
polyfills [Dynamo] Trace enter/exit of TorchFunctionModes (#135422) (#137114) 2024-10-09 02:29:40 +00:00
repro [logging/debugging] handle None (constant) args in debug log (#137032) 2024-10-02 01:43:22 +00:00
variables Revert "Introduce torch.sym_sum (#136429)" 2024-10-09 20:08:01 +00:00
__init__.py
_trace_wrapped_higher_order_op.py type _dynamo/trace_wrapped_higher_order_op.py (#137354) 2024-10-09 18:35:28 +00:00
bytecode_analysis.py
bytecode_transformation.py [dynamo] refactor resume_execution.py to use bytecode templates (#136483) 2024-09-24 22:20:26 +00:00
cache_size.py Avoid __ne__ weakref comparison and use identity instead in cache_size.py (#135000) 2024-10-03 07:43:58 +00:00
callback.py turn CompilationCallbackHandler into dataclass (#137312) 2024-10-05 19:03:28 +00:00
code_context.py add types to _dynamo/code_context.py (#136665) 2024-09-27 18:27:42 +00:00
codegen.py Correctly convert Python float to float64 when passing argument as Tensor (#136413) 2024-09-26 16:50:13 +00:00
compiled_autograd.py [compiled autograd] initialize cudagraph tls from context manager (#136735) 2024-10-03 06:22:11 +00:00
comptime.py
config.py [export] Custom op meta kernel generation (two pass) (#137277) 2024-10-07 15:34:19 +00:00
convert_frame.py [Dynamo] Trace enter/exit of TorchFunctionModes (#135422) (#137114) 2024-10-09 02:29:40 +00:00
create_parameter_op.py type _dynamo/create_parameter_op.py (#136958) 2024-10-02 10:23:37 +00:00
current_scope_id.py add typing to _dynamo/current_scope_id.py (#136676) 2024-09-27 04:09:15 +00:00
debug_utils.py [logging/debugging] handle None (constant) args in debug log (#137032) 2024-10-02 01:43:22 +00:00
decorators.py [dynamo] simplify implementation for functools.reduce (#133778) 2024-09-16 04:53:06 +00:00
device_interface.py Use torch.Stream&torch.Event for Dynamo capature (#134850) 2024-10-02 14:15:33 +00:00
distributed.py
eval_frame.py [dynamo] Log a summary of frames Dynamo traced (#137297) 2024-10-07 19:44:41 +00:00
exc.py [dynamo] attempt run only mode when dynamo cache limit is hit (#136655) 2024-09-27 17:15:05 +00:00
external_utils.py type _dynamo/external_utils.py (#137185) 2024-10-03 15:18:53 +00:00
funcname_cache.py
guards.py [Dynamo] Remove ignored modes from torch function mode stack guard (#135503) (#137116) 2024-10-09 02:29:40 +00:00
hooks.py
logging.py type _dynamo/logging.py (#136956) 2024-10-01 14:35:54 +00:00
mutation_guard.py type _dynamo/mutation_guard.py (#137350) 2024-10-08 00:04:34 +00:00
output_graph.py Revert "Introduce torch.sym_sum (#136429)" 2024-10-09 20:08:01 +00:00
profiler.py type _dynamo/profiler.py (#137351) 2024-10-07 18:54:33 +00:00
replay_record.py type _dynamo/replay_record.py (#137183) 2024-10-04 20:29:24 +00:00
resume_execution.py [Dynamo] Trace enter/exit of TorchFunctionModes (#135422) (#137114) 2024-10-09 02:29:40 +00:00
side_effects.py [Dynamo] Trace enter/exit of TorchFunctionModes (#135422) (#137114) 2024-10-09 02:29:40 +00:00
source.py Allow automatic dynamic shapes for closures and set current node properly in flexattention subgraph lowering (#137043) 2024-10-01 09:08:08 +00:00
symbolic_convert.py [dynamo] Fix error when inlining certain nested closure returned by another function (#137510) 2024-10-09 18:13:57 +00:00
tensor_version_op.py
test_case.py type _dynamo/test_case.py (#136957) 2024-10-01 14:36:22 +00:00
test_minifier_common.py
testing.py [Dynamo] Trace enter/exit of TorchFunctionModes (#135422) (#137114) 2024-10-09 02:29:40 +00:00
trace_rules.py [Dynamo] Handle torch function subclass/mode dispatch on generic tensor methods (#137119) 2024-10-09 02:29:40 +00:00
types.py Fix PT2 Source Code Annotations (#136460) 2024-09-28 03:54:43 +00:00
utils.py [Dynamo] Handle torch function subclass/mode dispatch on generic tensor methods (#137119) 2024-10-09 02:29:40 +00:00