Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/alibaba/BladeDISC
/ functions
Functions
6,088 in github.com/alibaba/BladeDISC
⨍
Functions
6,088
◇
Types & classes
1,279
↳
Endpoints
6
↓ 23 callers
Function
bitonicExchange
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:16
↓ 23 callers
Function
bitonicExchangePair
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:27
↓ 23 callers
Method
reorder_if_differ_in
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:654
↓ 22 callers
Function
StrSplit
pytorch_blade/pytorch_blade/common_utils/utils.cpp:72
↓ 22 callers
Method
alloc
tao_compiler/mlir/xla/ral/context/base/base_context.cc:110
↓ 22 callers
Method
insert
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/lru_cache.hpp:101
↓ 22 callers
Function
isRank2ColReduction
Returns true if this op is a rank-2 column reduction.
tao_compiler/mlir/disc/transforms/fusion_utils.cc:478
↓ 22 callers
Method
num_inputs
(self)
tensorflow_blade/tf_blade/util/simple_graph.py:42
↓ 22 callers
Function
swap
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:262
↓ 22 callers
Function
useShapeConstraintIR
tao_compiler/mlir/disc/disc_util.cc:199
↓ 21 callers
Method
_test_func
(self, torch_func, test_int=True)
pytorch_blade/tests/disc/ops/test_binary_ops.py:86
↓ 21 callers
Method
appendNode
(self, node)
pytorch_blade/torch_blade/onnx_backends/backend_testbed.py:355
↓ 21 callers
Method
g
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:418
↓ 21 callers
Function
isRank2RowReduction
Returns true if this op is a rank-2 row reduction.
tao_compiler/mlir/disc/transforms/fusion_utils.cc:449
↓ 21 callers
Method
mapSymbolicDimEqual
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:138
↓ 21 callers
Function
mayCreateStore
tao_compiler/mlir/disc/transforms/lhlo_elemental_utils.cc:123
↓ 21 callers
Function
permute_final_dims
(tensor, inds)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/modules/utils.py:52
↓ 20 callers
Function
DebugString
TODO(jpienaar): Move to OutputTensor.
tao/tao_bridge/passes/functionalize_cond.cc:49
↓ 20 callers
Method
DebugString
tao/tao_bridge/passes/tao_mark_for_compilation_pass.cc:225
↓ 20 callers
Method
GetSequence
tao/tao_bridge/tf/ordered_set.h:78
↓ 20 callers
Method
_test_e2e
( self, model, inp, pdll_files=None, pdll_dirs=None, enable_int8=False )
pytorch_blade/tests/disc/pdl/test_e2e/test_conv_bias.py:36
↓ 20 callers
Method
findLastWriter
Here `value` is supposed to be a pointer to buffer. Returns the defining op of `value `if no known op updates the buffer, otherwise returns the last o
tao_compiler/mlir/disc/transforms/fusion_utils.h:248
↓ 20 callers
Method
getDominantOp
Returns the dominant op of this fusion pattern. For kLoop fusion, a dominant op may be any op that has external users. For kInput fusion, a dominant o
tao_compiler/mlir/disc/transforms/fusion_utils.h:315
↓ 20 callers
Method
matmul
(x, y)
pytorch_blade/tests/disc/ops/test_matmul.py:71
↓ 19 callers
Function
DumpGraphToFile
tao/tao_bridge/tf/dump_graph.cc:96
↓ 19 callers
Function
GetXlaClusterForNode
tao/tao_bridge/tf/xla_cluster_util.cc:203
↓ 19 callers
Method
_test_single_layer
(self, model, inp)
pytorch_blade/tests/neural_engine/test_support_info.py:25
↓ 19 callers
Method
getOpContext
tao_compiler/mlir/xla/ral/context/tensorflow/tf_context_impl.cc:155
↓ 19 callers
Method
id
An ID assigned to the Predicate at construction time. Conceptually like a pointer, except that it is stable across runs.
tao/tao_bridge/tf/deadness_analysis.cc:130
↓ 19 callers
Function
optimize
Optimize the inference process of a PyTorch model. The whole model (or part of its submodules) will be exported to TorchScript to obtain its
pytorch_blade/torch_blade/optimization.py:88
↓ 19 callers
Function
to_bytes
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/utils.hpp:154
↓ 19 callers
Method
to_format_any
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:223
↓ 18 callers
Function
gather
(input: Tensor, dim: int = -1, dtype=torch.float)
examples/PyTorch/Inference/CUDA/AlphaFold/FoldAcc/foldacc/model/distributed/comm.py:68
↓ 18 callers
Function
getLhloOpsElementType
tao_compiler/mlir/disc/disc_util.cc:373
↓ 18 callers
Function
get_source_root_dir
()
pytorch_blade/common_setup.py:85
↓ 18 callers
Function
reportErrorIfAny
tao_compiler/mlir/xla/ral/context/base/cuda/cuda_context_impl.cc:95
↓ 18 callers
Method
run
tao_compiler/mlir/disc/transforms/shape_utils.cc:64
↓ 17 callers
Function
LogNotCompilable
tao/tao_bridge/tf/compilability_check_util.cc:74
↓ 17 callers
Method
func
(x, y)
pytorch_blade/tests/disc/ops/test_slices.py:97
↓ 17 callers
Method
getRootSymbolicDim
tao_compiler/mlir/disc/transforms/disc_shape_optimization_utils.cc:121
↓ 17 callers
Function
hash_combine
tao_compiler/mlir/xla/ral/context/common_context_impl_mkldnn.h:84
↓ 17 callers
Function
isGpuMemRef
Return true if the memref is on GPU
tao_compiler/mlir/disc/transforms/placement_utils.cc:54
↓ 17 callers
Function
to_string
tao_compiler/mlir/disc/utils/source_emitter.cc:29
↓ 16 callers
Method
GetNanoSeconds
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:145
↓ 16 callers
Function
GetTaoDumperOptions
tao/tao_bridge/dumper_common.cc:126
↓ 16 callers
Method
_check_type
(self, out, res)
pytorch_blade/tests/disc/ops/test_binary_ops.py:20
↓ 16 callers
Method
_test_reshape
(self, reshape_func, dtype=None, x=None)
pytorch_blade/tests/disc/ops/test_shapes.py:20
↓ 16 callers
Function
cleanUnusedLhloOps
tao_compiler/mlir/disc/transforms/fusion_utils.cc:109
↓ 16 callers
Function
condExchange
tao_compiler/mlir/xla/ral/context/custom_library/bitonic_sort.cu.h:269
↓ 16 callers
Function
getNumResultOperands
tao_compiler/mlir/disc/disc_util.cc:353
↓ 16 callers
Method
getOrCreate
tao_compiler/mlir/xla/ral/context/common_context_impl_mkldnn.h:796
↓ 16 callers
Method
get_dim
Return size of specified dimension
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:596
↓ 16 callers
Method
num_outputs
(self)
tensorflow_blade/tf_blade/util/simple_graph.py:46
↓ 15 callers
Method
CheckInvariants
tao/tao_bridge/tf/graphcycles.cc:88
↓ 15 callers
Method
Stop
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:129
↓ 15 callers
Function
_jit_pass_clean_python_ir
(graph, is_training=False)
pytorch_blade/torch_blade/python_ir_analysis.py:252
↓ 15 callers
Method
_test_nn_ops
(self, nn_ops_func, x=None)
pytorch_blade/tests/disc/ops/test_nn_ops.py:23
↓ 15 callers
Function
buildTileOp
tao_compiler/mlir/disc/transforms/disc_transform_schedule.cc:138
↓ 15 callers
Function
calcLinearIndex
tao_compiler/mlir/disc/transforms/codegen_utils.cc:111
↓ 15 callers
Function
calcMultiDimIndex
tao_compiler/mlir/disc/transforms/codegen_utils.cc:118
↓ 15 callers
Function
cwd
Change the current working directory to `path` to do somthing and then recover the current cwd when it's done.
scripts/python/common_setup.py:185
↓ 15 callers
Method
dump
tao_compiler/mlir/disc/transforms/shape_utils.h:111
↓ 15 callers
Function
execute
Execute a shell command, exception raised on failure.
scripts/python/common_setup.py:146
↓ 15 callers
Function
fx_trace
( root: nn.Module, module_filter: Optional[ModuleFilter] = None, tracer: Optional[Tracer] = None,
tools/torch_quant/torch_quant/module.py:124
↓ 15 callers
Method
has_scale
Return whether the param has a scale
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:898
↓ 15 callers
Function
which
Same as `which` command of bash.
pytorch_blade/common_setup.py:174
↓ 14 callers
Method
RemoveNode
tao/tao_bridge/tf/graphcycles.cc:128
↓ 14 callers
Method
_test_activation
(self, nn_func=None, native_func=None, dims=[])
pytorch_blade/tests/disc/ops/test_activation.py:21
↓ 14 callers
Function
createParallelAndSetInsPt
tao_compiler/mlir/disc/transforms/codegen_utils.cc:162
↓ 14 callers
Method
d2h
tao_compiler/mlir/xla/ral/device/gpu/gpu_driver.cc:164
↓ 14 callers
Method
decode
decodes input (unsigned char) into output (uint4). Assumes len is a multiple of 4.
tao_compiler/mlir/xla/ral/ral_md5.cc:147
↓ 14 callers
Method
device_type
tao/tao_bridge/kernels/platform_info.h:35
↓ 14 callers
Function
execute
Execute a shell command, exception raised on failure.
scripts/python/tao_common.py:57
↓ 14 callers
Method
graph
pytorch_blade/pytorch_blade/ltc/disc_compiler/disc_compiler.h:44
↓ 14 callers
Function
isMemIntensiveOptExperimentalEnabled
tao_compiler/mlir/disc/disc_util.cc:253
↓ 14 callers
Function
isTorchMlirEnable
()
pytorch_blade/tests/disc/testing_base.py:35
↓ 14 callers
Function
set_attr_int
(node: tf.NodeDef, key: str, value: int)
tensorflow_blade/tf_blade/util/tf_graph_transform_util.py:127
↓ 14 callers
Function
truncateShuffleElemType
tao_compiler/mlir/disc/transforms/lhlo_legalize_roots_to_loops.cc:1288
↓ 14 callers
Function
update
(args)
scripts/python/bazel_snapshot_client.py:102
↓ 13 callers
Method
GetOperands
tao/tao_bridge/tf/deadness_analysis.cc:202
↓ 13 callers
Method
_test_slice
(self, reshape_func, dtype=None, x=None)
pytorch_blade/tests/disc/ops/test_slices.py:21
↓ 13 callers
Method
_test_torchscipte_to_mhlo
(self, graph, expected_str)
pytorch_blade/tests/disc/testing_base.py:105
↓ 13 callers
Method
bindInput
tao_compiler/mlir/xla/ral/context/base/base_context.cc:63
↓ 13 callers
Method
clone
(self)
pytorch_blade/torch_blade/config.py:125
↓ 13 callers
Function
emitString
tao_compiler/mlir/disc/transforms/disc_lower_to_library_call.cc:692
↓ 13 callers
Function
getValueUsers
Returns data users of the value and its aliases (e.g. memref.cast). Here non-data users means DimOp, DeallocOp and ShapeOfOp.
tao_compiler/mlir/disc/disc_util.cc:285
↓ 13 callers
Function
isProfilingEnabled
tao_compiler/mlir/xla/ral/context/common_context_impl.cc:116
↓ 13 callers
Method
is_nhwc
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:154
↓ 13 callers
Function
parseStr
tao_compiler/mlir/xla/ral/context/pdll_util.cc:26
↓ 13 callers
Method
quantize
(self, model: nn.Module)
tools/torch_quant/torch_quant/quantizer.py:256
↓ 13 callers
Method
rank
tao_compiler/mlir/disc/transforms/shape_utils.h:48
↓ 13 callers
Function
tensor_zp_mask
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/utils.hpp:268
↓ 13 callers
Method
to_format
tao_compiler/mlir/xla/ral/context/mkldnn/ideep/ideep/tensor.hpp:749
↓ 13 callers
Method
trace
(self, root, **kwargs)
examples/PyTorch/Quantization/bert_ptq.py:176
↓ 13 callers
Method
update
MD5 block update operation. Continues an MD5 message-digest operation, processing another message block
tao_compiler/mlir/xla/ral/ral_md5.cc:258
↓ 13 callers
Function
updateAssignment
tao_compiler/mlir/disc/transforms/disc_assign_memory_space.cc:61
↓ 12 callers
Function
IsInSubgraph
tao/tao_bridge/passes/tao_encapsulate_subgraphs_pass.cc:1471
↓ 12 callers
Method
RemoveEdge
tao/tao_bridge/tf/graphcycles.cc:153
↓ 12 callers
Function
TEST
tao_compiler/mlir/disc/tests/regression/onednn.cc:29
↓ 12 callers
Method
_test_binary_type_promotion
(self, torch_func)
pytorch_blade/tests/disc/ops/test_binary_ops.py:32
← previous
next →
101–200 of 6,088, ranked by callers