..
dbr_quantization
dbr quant: add torchscript pass to remove redundant aliases ( #71230 )
2022-03-03 15:31:53 +00:00
onnx
[ONNX] Add setType from user into InferredType and Reliable in ConstantValueMap ( #88622 )
2022-11-19 17:16:59 +00:00
quantization
quantization: deprecate observer compute_dtype and replace with is_dynamic ( #85431 )
2022-11-24 07:07:34 +00:00
utils
Avoid unnecessary copy of pointeeSet in MemoryDAG::setWildcards ( #83681 )
2022-08-19 14:14:32 +00:00
add_if_then_else.cpp
Replace all CHECK_ and DCHECK_ with TORCH_* macros ( #82032 )
2022-07-26 01:20:44 +00:00
add_if_then_else.h
annotate_warns.cpp
annotate_warns.h
autocast.cpp
[JIT] Retry - Support scripting torch.is_autocast_enabled() ( #82394 )
2022-08-10 18:26:17 +00:00
autocast.h
bailout_graph.cpp
bailout_graph.h
batch_mm.cpp
batch_mm.h
canonicalize_graph_fuser_ops.cpp
canonicalize_graph_fuser_ops.h
canonicalize.cpp
canonicalize.h
check_strict_fusion.cpp
[JIT] Add Context Manager to force strict fusion
2022-04-25 16:08:57 +00:00
check_strict_fusion.h
[JIT] Add Context Manager to force strict fusion
2022-04-25 16:08:57 +00:00
clear_profiling.cpp
clear_profiling.h
clear_undefinedness.cpp
clear_undefinedness.h
common_subexpression_elimination.cpp
Revert "Dnt CSE across context managers"
2022-04-22 20:44:57 +00:00
common_subexpression_elimination.h
concat_opt.cpp
concat_opt.h
constant_pooling.cpp
constant_pooling.h
constant_propagation.cpp
constant_propagation.h
create_autodiff_subgraphs.cpp
Retry - [JIT] Propagate profiled information to DifferentiableGraph outputs
2022-06-21 22:57:17 +00:00
create_autodiff_subgraphs.h
create_functional_graphs.cpp
create_functional_graphs.h
cuda_graph_fuser.cpp
[JIT] Optionally validate nvfuser outputs after execution ( #74361 )
2022-04-01 23:48:30 +00:00
cuda_graph_fuser.h
[NVFuser] environment variable to turn nvfuser on or off ( #76485 )
2022-05-03 23:05:40 +00:00
dead_code_elimination.cpp
dead_code_elimination.h
decompose_ops.cpp
adjust conditions to enable jit decomposition pass only for GPU device ( #73637 )
2022-03-21 21:39:33 +00:00
decompose_ops.h
device_type_analysis.cpp
device_type_analysis.h
dtype_analysis.cpp
dtype_analysis.h
eliminate_no_ops.cpp
eliminate_no_ops.h
erase_number_types.cpp
erase_number_types.h
fixup_trace_scope_blocks.cpp
fixup_trace_scope_blocks.h
fold_conv_bn.cpp
New generated conv_bn folding should use same weight and bias dtype as original conv module ( #77042 )
2022-05-09 15:17:57 +00:00
fold_conv_bn.h
freeze_module.cpp
[JIT] Allow freezing modules that contain mutable interfaces ( #86039 )
2022-10-08 00:38:11 +00:00
freeze_module.h
frozen_concat_linear.cpp
frozen_concat_linear.h
frozen_conv_add_relu_fusion_cuda.cpp
[ROCm] MIOpen fused convolution relu ( #82002 )
2022-08-16 20:49:33 +00:00
frozen_conv_add_relu_fusion.cpp
frozen_conv_add_relu_fusion.h
frozen_conv_folding.cpp
Conv BN folding data type issue when conv has no bias ( #78241 )
2022-05-26 18:42:17 +00:00
frozen_conv_folding.h
expand the coverage of conv folding ( #75724 )
2022-05-11 14:29:16 +00:00
frozen_graph_optimizations.cpp
expand the coverage of conv folding ( #75724 )
2022-05-11 14:29:16 +00:00
frozen_graph_optimizations.h
frozen_linear_transpose.cpp
frozen_linear_transpose.h
frozen_ops_to_mkldnn.cpp
set the correct size of aten tensor in presence of mkldnn padding ( #86767 )
2022-10-13 19:35:31 +00:00
frozen_ops_to_mkldnn.h
fuse_linear.cpp
fuse_linear.h
fuse_relu.cpp
fuse_relu.h
graph_fuser.cpp
Fix some typos.
2022-04-11 21:55:59 +00:00
graph_fuser.h
graph_rewrite_helper.cpp
extend replaceConvolutionWithAtenConv to handle conv_transpose3d ( #76888 )
2022-05-13 16:40:12 +00:00
graph_rewrite_helper.h
guard_elimination.cpp
guard_elimination.h
hoist_conv_packed_params.cpp
[quant][ao_migration] torch.nn.quantized.modules → torch.ao.nn.quantized.modules ( #78713 )
2022-08-25 16:50:33 +00:00
hoist_conv_packed_params.h
inline_autodiff_subgraphs.cpp
[Reland take-2] Add JIT graph fuser for oneDNN Graph API (v0.5)
2022-05-05 16:57:03 +00:00
inline_autodiff_subgraphs.h
inline_fork_wait.cpp
inline_fork_wait.h
inline_forked_closures.cpp
inline_forked_closures.h
inliner.cpp
Add api for recursively analyzing function calls ( #73329 )
2022-02-28 17:44:45 +00:00
inliner.h
Add api for recursively analyzing function calls ( #73329 )
2022-02-28 17:44:45 +00:00
inplace_check.cpp
inplace_check.h
insert_guards.cpp
insert_guards.h
integer_value_refinement.cpp
integer_value_refinement.h
lift_closures.cpp
lift_closures.h
liveness.cpp
liveness.h
loop_unrolling.cpp
loop_unrolling.h
lower_grad_of.cpp
fix to map an undefined tensor back to a tensor list
2022-04-07 20:07:27 +00:00
lower_grad_of.h
lower_graph.cpp
lower_graph.h
lower_tuples.cpp
lower_tuples.h
metal_rewrite.cpp
metal_rewrite.h
mkldnn_rewrite.cpp
[NNC] support aten::_convolution when it is 2D conv ( #84038 )
2022-09-19 17:45:20 +00:00
mkldnn_rewrite.h
[NNC] enable fusion of conv with elementwise OP ( #77157 )
2022-08-10 21:46:51 +00:00
mobile_optimizer_type.h
[Vulkan] Add Vulkan Rewrite to Transfer Inputs and Outputs to Vulkan and CPU Backends Respectively ( #87432 )
2022-10-31 14:18:45 +00:00
normalize_ops.cpp
[reland] [numpy] add torch.concatenate, alias of torch.cat ( #85073 )
2022-09-15 19:34:44 +00:00
normalize_ops.h
onednn_graph_fuser.h
[Reland take-2] Add JIT graph fuser for oneDNN Graph API (v0.5)
2022-05-05 16:57:03 +00:00
onnx.cpp
[ONNX] Move two headers from .h to .cc ( #86852 )
2022-11-24 01:30:09 +00:00
onnx.h
[ONNX] Move two headers from .h to .cc ( #86852 )
2022-11-24 01:30:09 +00:00
pass_manager.cpp
pass_manager.h
[JIT] C10_EXPORT -> TORCH_API ( #73818 )
2022-03-14 20:29:58 +00:00
peephole_alias_sensitive.cpp
peephole_alias_sensitive.h
peephole_dict_idioms.cpp
peephole_dict_idioms.h
peephole_list_idioms.cpp
peephole_list_idioms.h
peephole_non_tensor.cpp
Fix typos in messages under torch ( #88961 )
2022-11-14 19:06:41 +00:00
peephole_non_tensor.h
peephole.cpp
peephole.h
prepack_folding.cpp
prepack_folding.h
refine_tuple_types.cpp
Add RefineTypes JIT pass for Tuple ( #76919 )
2022-05-12 00:48:39 +00:00
refine_tuple_types.h
Add RefineTypes JIT pass for Tuple ( #76919 )
2022-05-12 00:48:39 +00:00
remove_dropout.cpp
remove_dropout.h
remove_exceptions.cpp
remove_exceptions.h
remove_expands.cpp
remove_expands.h
remove_inplace_ops.cpp
remove_inplace_ops.h
remove_mutation.cpp
remove_mutation.h
remove_redundant_profiles.cpp
remove_redundant_profiles.h
replacement_of_old_operators.cpp
Remove -Wno-unused-variable from utils.cmake (take 2) ( #75538 )
2022-04-20 17:41:59 +00:00
replacement_of_old_operators.h
requires_grad_analysis.cpp
requires_grad_analysis.h
restore_mutation.cpp
restore_mutation.h
shape_analysis.cpp
Lower randint default dtype to the C++ API ( #81410 )
2022-07-21 16:42:49 +00:00
shape_analysis.h
specialize_autogradzero.cpp
Clean up profiling mode and profiling executor strategy ( #73875 )
2022-03-29 18:38:51 +00:00
specialize_autogradzero.h
subgraph_rewrite.cpp
subgraph_rewrite.h
symbolic_shape_analysis.cpp
Symbolic Shape Analaysis: Add Generalized List of Tensor Shape Support ( #78679 )
2022-08-17 19:13:26 +00:00
symbolic_shape_analysis.h
Relanding shape cache (75400) ( #75710 )
2022-04-13 17:17:30 +00:00
symbolic_shape_cache.cpp
Relanding shape cache (75400) ( #75710 )
2022-04-13 17:17:30 +00:00
symbolic_shape_cache.h
Relanding shape cache (75400) ( #75710 )
2022-04-13 17:17:30 +00:00
symbolic_shape_runtime_fusion.cpp
[NNC] channels last propagation within NNC fusion group ( #76948 )
2022-05-30 18:31:49 +00:00
symbolic_shape_runtime_fusion.h
tensorexpr_fuser.cpp
Optimize to if the datatyep of the source tensor is as same as the dest datatype ( #85140 )
2022-09-27 04:40:42 +00:00
tensorexpr_fuser.h
Moving Remove Tensor Type Specializations to after custom passes
2022-04-11 22:12:01 +00:00
update_differentiable_graph_requires_grad.cpp
update_differentiable_graph_requires_grad.h
value_refinement_utils.cpp
value_refinement_utils.h
variadic_ops.cpp
variadic_ops.h
vulkan_rewrite.cpp
[vulkan] enable prepacking for Batchnorm op ( #88433 )
2022-11-04 19:24:13 +00:00
vulkan_rewrite.h
[Vulkan] Add support for Optimization Blocklist to Vulkan Rewrite ( #87431 )
2022-10-31 14:15:51 +00:00
xnnpack_rewrite.cpp
[Vulkan] Add support for Optimization Blocklist to Vulkan Rewrite ( #87431 )
2022-10-31 14:15:51 +00:00
xnnpack_rewrite.h
[Vulkan] Add support for Optimization Blocklist to Vulkan Rewrite ( #87431 )
2022-10-31 14:15:51 +00:00
