p60975123
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

!14737 add is dyanmic format op info 

From: @jjfeing
Reviewed-by: @kisnwang,@zhoufeng54
Signed-off-by: @zhoufeng54
This commit is contained in:
mindspore-ci-bot 2021-04-23 15:03:59 +08:00 committed by Gitee
parent b77c3f093d 60c95ea004
commit 181c4d31c0
44 changed files with 80 additions and 72 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
mindspore/ccsrc/backend/kernel_compiler/kernel.h
View File

@ -53,7 +53,6 @@ enum OpPattern {
kFormatAgnosticPattern = 1,
kBroadcastPattern = 2,
kReducePattern = 3,
kDynamicFormatPattern = 4,
};
// Backend processor

8
mindspore/ccsrc/backend/kernel_compiler/oplib/opinfo.h
View File

@ -101,11 +101,11 @@ class OpInfo {
compute_cost_ = opinfo.compute_cost_;
kernel_name_ = opinfo.kernel_name();
partial_flag_ = opinfo.partial_flag_;
dynamic_format_ = opinfo.dynamic_format_;
dynamic_shape_ = opinfo.dynamic_shape_;
op_pattern_ = opinfo.op_pattern();
processor_ = opinfo.processor_;
need_check_supported_ = opinfo.need_check_supported();
is_dynamic_format_ = opinfo.is_dynamic_format();
for (const auto &attr : opinfo.attrs_ptr()) {
attrs_ptr_.push_back(std::make_shared<OpAttr>(*attr));
}
@ -127,6 +127,7 @@ class OpInfo {
bool dynamic_shape() const { return dynamic_shape_; }
std::string processor() const { return processor_; }
bool need_check_supported() const { return need_check_supported_; }
bool is_dynamic_format() const { return is_dynamic_format_; }
std::vector<std::shared_ptr<OpAttr>> attrs_ptr() const { return attrs_ptr_; }
std::vector<std::shared_ptr<OpIOInfo>> inputs_ptr() const { return inputs_ptr_; }
std::vector<std::shared_ptr<OpIOInfo>> outputs_ptr() const { return outputs_ptr_; }
@ -144,7 +145,8 @@ class OpInfo {
void set_partial_flag(const bool partial_flag) { partial_flag_ = partial_flag; }
void set_op_pattern(const OpPattern op_pattern) { op_pattern_ = op_pattern; }
void set_processor(const std::string &processor) { processor_ = processor; }
void set_need_check_supported(const bool need_check_supported) { need_check_supported_ = need_check_supported; }
void set_need_check_supported(bool need_check_supported) { need_check_supported_ = need_check_supported; }
void set_is_dynamic_format(bool is_dynamic_format) { is_dynamic_format_ = is_dynamic_format; }
void add_attrs_ptr(const std::shared_ptr<OpAttr> &attr) { attrs_ptr_.push_back(attr); }
void add_inputs_ptr(const std::shared_ptr<OpIOInfo> &input) { inputs_ptr_.push_back(input); }
void add_outputs_ptr(const std::shared_ptr<OpIOInfo> &output) { outputs_ptr_.push_back(output); }
@ -169,9 +171,9 @@ class OpInfo {
int compute_cost_ = 0;
std::string kernel_name_;
bool partial_flag_ = false;
bool dynamic_format_ = false;
bool dynamic_shape_ = false;
bool need_check_supported_ = false;
bool is_dynamic_format_ = false;
OpPattern op_pattern_ = kCommonPattern;
std::string processor_;
std::vector<std::shared_ptr<OpAttr>> attrs_ptr_;

11
mindspore/ccsrc/backend/kernel_compiler/oplib/oplib.cc
View File

@ -34,6 +34,7 @@ constexpr auto kKernelName = "kernel_name";
constexpr auto kPartialFlag = "partial_flag";
constexpr auto kReshapeType = "reshape_type";
constexpr auto kOpPattern = "op_pattern";
constexpr auto kIsDynamicFormat = "is_dynamic_format";
constexpr auto kDynamicFormat = "dynamicFormat";
constexpr auto kFormatAgnostic = "formatAgnostic";
constexpr auto kNeedCheckSupported = "need_check_supported";
@ -102,10 +103,8 @@ bool OpLib::RegOp(const std::string &json_string, const std::string &impl_path)
}
void OpLib::DecodeTBESpecificInfo(const nlohmann::json &obj, const std::shared_ptr<OpInfo> &op_info) {
const std::map<std::string, kernel::OpPattern> kOpPatternMap = {{kFormatAgnostic, kFormatAgnosticPattern},
{kBroadcast, kBroadcastPattern},
{kReduce, kReducePattern},
{kDynamicFormat, kDynamicFormatPattern}};
const std::map<std::string, kernel::OpPattern> kOpPatternMap = {
{kFormatAgnostic, kFormatAgnosticPattern}, {kBroadcast, kBroadcastPattern}, {kReduce, kReducePattern}};
MS_EXCEPTION_IF_NULL(op_info);
op_info->set_async_flag(obj.at(kAsyncFlag));
op_info->set_binfile_name(obj.at(kBinfileName));
@ -118,6 +117,10 @@ void OpLib::DecodeTBESpecificInfo(const nlohmann::json &obj, const std::shared_p
op_info->set_dynamic_shape(obj.at(kDynamicShape));
}
if (obj.find(kIsDynamicFormat) != obj.end()) {
op_info->set_is_dynamic_format(obj.at(kIsDynamicFormat));
}
if (obj.find(kOpPattern) != obj.end()) {
std::string op_pattern = obj.at(kOpPattern);
auto find_iter = kOpPatternMap.find(op_pattern);

7
mindspore/ccsrc/backend/kernel_compiler/tbe/tbe_kernel_select/tbe_kernel_select.cc
View File

@ -63,11 +63,13 @@ void TbeKernelSelect::TbeMetadataInfoEx() {
MS_LOG(INFO) << "Warning: node(" << cnode_ptr_->fullname_with_scope() << ") not support tbe aicore.";
return;
}
if (op_info_ptr->is_dynamic_format()) {
GetDynamicFormatPatternKernelInfo(*op_info_ptr);
} else {
OpPattern pattern = op_info_ptr->op_pattern();
if (pattern == kCommonPattern) {
GetCommonPatternKernelInfo(*op_info_ptr);
} else if (pattern == kDynamicFormatPattern) {
GetDynamicFormatPatternKernelInfo(*op_info_ptr);
} else if (pattern == kFormatAgnosticPattern) {
GetAgnosticPatternKernelInfo(*op_info_ptr);
} else if (pattern == kBroadcastPattern) {
@ -77,6 +79,7 @@ void TbeKernelSelect::TbeMetadataInfoEx() {
} else {
MS_LOG(INFO) << "Warning: op pattern is invailed.";
}
}
// check support
FilterInVaildKernelInfo(*op_info_ptr);
}

2
mindspore/ops/_op_impl/tbe/adam_apply_one.py
View File

@ -36,7 +36,7 @@ adam_apply_one_op_info = TBERegOp("AdamApplyOne") \
.output(0, "output0", False, "required", "all") \
.output(1, "output1", False, "required", "all") \
.output(2, "output2", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None, DataType.None_None, DataType.None_None,
DataType.None_None, DataType.None_None, DataType.None_None, DataType.None_None,
DataType.None_None, DataType.None_None, DataType.None_None, DataType.None_None,

2
mindspore/ops/_op_impl/tbe/batch_matmul.py
View File

@ -30,7 +30,7 @@ batch_matmul_op_info = TBERegOp("BatchMatMul") \
.input(1, "x2", False, "required", "all") \
.input(2, "bias", False, "optional", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.I32_None, DataType.I32_None, DataType.I32_None, DataType.I32_None) \
.dtype_format(DataType.F16_None, DataType.F16_None, DataType.F16_None, DataType.F16_None) \
.dtype_format(DataType.F32_None, DataType.F32_None, DataType.F32_None, DataType.F32_None) \

2
mindspore/ops/_op_impl/tbe/bce_with_logits_loss.py
View File

@ -23,7 +23,7 @@ bce_with_logits_loss_op_info = TBERegOp("BCEWithLogitsLoss") \
.compute_cost(10) \
.kernel_name("sigmoid_cross_entropy_with_logits_v2") \
.partial_flag(True) \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.attr("reduction", "optional", "str", "all", "mean") \
.input(0, "predict", False, "required", "all") \
.input(1, "target", False, "required", "all") \

2
mindspore/ops/_op_impl/tbe/bias_add.py
View File

@ -27,7 +27,7 @@ bias_add_grad_op_info = TBERegOp("BiasAdd") \
.input(0, "x", False, "required", "all") \
.input(1, "bias", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/binary_cross_entropy.py
View File

@ -28,7 +28,7 @@ binary_cross_entropy_op_info = TBERegOp("BinaryCrossEntropy") \
.input(1, "y", False, "required", "all") \
.input(2, "weight", False, "optional", "all") \
.output(0, "output", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None, DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/bn_training_reduce.py
View File

@ -26,7 +26,7 @@ bn_training_reduce_op_info = TBERegOp("BNTrainingReduce") \
.input(0, "x", False, "required", "all", reshape_type="NC") \
.output(0, "sum", False, "required", "all") \
.output(1, "square_sum", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.F16_None, DataType.F32_None, DataType.F32_None) \
.dtype_format(DataType.F32_None, DataType.F32_None, DataType.F32_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/bn_training_reduce_grad.py
View File

@ -32,7 +32,7 @@ bn_training_reduce_grad_op_info = TBERegOp("BNTrainingReduceGrad") \
.input(5, "batch_mean", False, "required", "all") \
.input(6, "batch_variance", False, "required", "all") \
.output(0, "y", False, "required", "all", reshape_type="NC") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.F16_None, DataType.F32_None, DataType.F32_None, DataType.F32_None,
DataType.F32_None, DataType.F32_None, DataType.F32_None, DataType.F16_None) \
.dtype_format(DataType.F32_None, DataType.F32_None, DataType.F32_None, DataType.F32_None,

2
mindspore/ops/_op_impl/tbe/bn_training_update_grad.py
View File

@ -30,7 +30,7 @@ bn_training_update_grad_op_info = TBERegOp("BNTrainingUpdateGrad") \
.input(3, "batch_variance", False, "required", "all") \
.output(0, "diff_scale", False, "required", "all") \
.output(1, "diff_offset", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.F16_None, DataType.F16_None, DataType.F32_None, DataType.F32_None,
DataType.F32_None, DataType.F32_None) \
.dtype_format(DataType.F32_None, DataType.F32_None, DataType.F32_None, DataType.F32_None,

2
mindspore/ops/_op_impl/tbe/bn_training_update_v2.py
View File

@ -32,7 +32,7 @@ bn_training_update_v2_op_info = TBERegOp("BNTrainingUpdateV2") \
.output(0, "y", False, "required", "all", reshape_type="NC") \
.output(1, "batch_mean", False, "required", "all") \
.output(2, "batch_variance", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.F16_None, DataType.F32_None, DataType.F32_None,
DataType.F32_None, DataType.F32_None, DataType.F16_None,
DataType.F32_None, DataType.F32_None) \

2
mindspore/ops/_op_impl/tbe/concat.py
View File

@ -26,7 +26,7 @@ concat_op_info = TBERegOp("Concat") \
.attr("axis", "required", "int", "all") \
.input(0, "input_values", False, "dynamic", "all") \
.output(0, "output_data", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/concat_ds.py
View File

@ -27,7 +27,7 @@ concat_ds_op_info = TBERegOp("Concat") \
.attr("axis", "required", "int", "all") \
.input(0, "input_values", False, "dynamic", "all") \
.output(0, "output_data", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/confusion_transpose_d.py
View File

@ -28,7 +28,7 @@ confusion_transpose_d_op_info = TBERegOp("ConfusionTransposeD") \
.attr("transpose_first", "required", "bool", "all") \
.input(0, "x", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

4
mindspore/ops/_op_impl/tbe/conv2d.py
View File

@ -23,7 +23,7 @@ conv2d_op_info = TBERegOp("Conv2D") \
.compute_cost(10) \
.kernel_name("conv2d") \
.partial_flag(True) \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.attr("stride", "required", "listInt", "all") \
.attr("pad_list", "required", "listInt", "all") \
.attr("dilation", "required", "listInt", "all") \
@ -35,7 +35,7 @@ conv2d_op_info = TBERegOp("Conv2D") \
.input(2, "bias", False, "optional", "all") \
.input(3, "offset_w", False, "optional", "all") \
.output(0, "y", True, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.F16_None, DataType.F16_None, DataType.F16_None, DataType.I8_None, DataType.F16_None) \
.dtype_format(DataType.I8_None, DataType.I8_None, DataType.I32_None, DataType.I8_None, DataType.I32_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/dropout_do_mask.py
View File

@ -27,7 +27,7 @@ drop_out_do_mask_op_info = TBERegOp("DropoutDoMask") \
.input(1, "mask", False, "required", "all") \
.input(2, "keep_prob", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None, DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/fused_mul_add.py
View File

@ -27,7 +27,7 @@ fused_mul_add_op_info = TBERegOp("FusedMulAdd") \
.input(1, "x2", False, "required", "all") \
.input(2, "x3", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None, DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/l2_loss.py
View File

@ -25,7 +25,7 @@ l2_loss_op_info = TBERegOp("L2Loss") \
.partial_flag(True) \
.input(0, "x", None, "required", None) \
.output(0, "y", True, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/layer_norm.py
View File

@ -32,7 +32,7 @@ layer_norm_op_info = TBERegOp("LayerNorm") \
.output(0, "y", False, "required", "all") \
.output(1, "mean", False, "required", "all") \
.output(2, "variance", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.F16_None, DataType.F16_None, DataType.F16_None, DataType.F16_None,
DataType.F16_None, DataType.F16_None) \
.dtype_format(DataType.F32_None, DataType.F32_None, DataType.F32_None, DataType.F32_None,

2
mindspore/ops/_op_impl/tbe/layer_norm_beta_gamma_backprop.py
View File

@ -30,7 +30,7 @@ layer_norm_beta_gamma_backprop_op_info = TBERegOp("LayerNormBetaGammaBackprop")
.input(3, "mean", False, "required", "all") \
.output(0, "pd_gamma", False, "required", "all") \
.output(1, "pd_beta", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.F16_None, DataType.F16_None, DataType.F16_None, DataType.F16_None,
DataType.F32_None, DataType.F32_None) \
.dtype_format(DataType.F32_None, DataType.F32_None, DataType.F32_None, DataType.F32_None,

2
mindspore/ops/_op_impl/tbe/layer_norm_x_backprop.py
View File

@ -29,7 +29,7 @@ layer_norm_x_backprop_op_info = TBERegOp("LayerNormXBackprop") \
.input(3, "mean", False, "required", "all") \
.input(4, "gamma", False, "required", "all") \
.output(0, "pd_x", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.F16_None, DataType.F16_None, DataType.F16_None, DataType.F16_None,
DataType.F16_None, DataType.F16_None) \
.dtype_format(DataType.F32_None, DataType.F32_None, DataType.F32_None, DataType.F32_None,

2
mindspore/ops/_op_impl/tbe/max_pool3d.py
View File

@ -33,7 +33,7 @@ max_pool3d_op_info = TBERegOp("MaxPool3D") \
.attr("format", "optional", "str", "all") \
.input(0, "x", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/mul.py
View File

@ -26,7 +26,7 @@ mul_op_info = TBERegOp("Mul") \
.input(0, "x", False, "required", "all") \
.input(1, "y", False, "required", "all") \
.output(0, "output", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/mul_ds.py
View File

@ -27,7 +27,7 @@ mul_ds_op_info = TBERegOp("Mul") \
.input(0, "x", False, "required", "all") \
.input(1, "y", False, "required", "all") \
.output(0, "output", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/prelu.py
View File

@ -26,7 +26,7 @@ prelu_op_info = TBERegOp("PReLU") \
.input(0, "x", False, "required", "all") \
.input(1, "weight", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/prelu_grad.py
View File

@ -28,7 +28,7 @@ prelu_grad_op_info = TBERegOp("PReLUGrad") \
.input(2, "weights", False, "required", "all") \
.output(0, "dx", False, "required", "all") \
.output(1, "da", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None, DataType.None_None, DataType.None_None,
DataType.None_None) \
.get_op_info()

1
mindspore/ops/_op_impl/tbe/real_div.py
View File

@ -27,6 +27,7 @@ realdiv_op_info = TBERegOp("RealDiv") \
.input(1, "y", False, "required", "all") \
.output(0, "z", False, "required", "all") \
.op_pattern("broadcast") \
.is_dynamic_format(True) \
.dtype_format(DataType.F16_None, DataType.F16_None, DataType.F16_None) \
.dtype_format(DataType.F32_None, DataType.F32_None, DataType.F32_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/reciprocal.py
View File

@ -25,7 +25,7 @@ reciprocal_op_info = TBERegOp("Reciprocal") \
.partial_flag(True) \
.input(0, "x", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/reverse_v2_d.py
View File

@ -23,7 +23,7 @@ reverse_v2_d_op_info = TBERegOp("ReverseV2") \
.compute_cost(10) \
.kernel_name("reverse_v2_d") \
.partial_flag(True) \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.attr("axis", "required", "listInt", "all") \
.input(0, "x", False, "required", "all") \
.output(0, "y", False, "required", "all") \

2
mindspore/ops/_op_impl/tbe/select.py
View File

@ -27,7 +27,7 @@ select_op_info = TBERegOp("Select") \
.input(1, "x1", False, "required", "all") \
.input(2, "x2", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None, DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/slice.py
View File

@ -23,7 +23,7 @@ slice_op_info = TBERegOp("Slice") \
.compute_cost(10) \
.kernel_name("slice_d") \
.partial_flag(True) \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.attr("begin", "required", "listInt", "all") \
.attr("size", "required", "listInt", "all") \
.input(0, "x", False, "required", "all") \

2
mindspore/ops/_op_impl/tbe/softmax.py
View File

@ -26,7 +26,7 @@ softmax_op_info = TBERegOp("Softmax") \
.attr("axis", "optional", "listInt", "all") \
.input(0, "x", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

4
mindspore/ops/_op_impl/tbe/softmax_grad_ext.py
View File

@ -23,14 +23,14 @@ softmax_grad_ext_op_info = TBERegOp("SoftmaxGradExt") \
.compute_cost(10) \
.kernel_name("softmax_grad_ext") \
.partial_flag(True) \
.dynamic_format(True) \
.is_dynamic_format(True) \
.attr("axis", "required", "listInt", "all") \
.attr("keepdims", "required", "bool", "all") \
.input(0, "grad", False, "required", "all") \
.input(1, "x1", False, "required", "all") \
.input(2, "x2", False, "required", "all") \
.output(0, "y", True, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None,
DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/split_d.py
View File

@ -27,7 +27,7 @@ split_d_op_info = TBERegOp("Split") \
.attr("output_num", "required", "int", "all") \
.input(0, "value", False, "required", "all") \
.output(0, "output", False, "dynamic", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/splitv.py
View File

@ -28,7 +28,7 @@ split_v_op_info = TBERegOp("SplitV") \
.attr("num_split", "required", "int", "all") \
.input(0, "x", False, "required", "all") \
.output(0, "y", False, "dynamic", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/strided_slice_d.py
View File

@ -33,7 +33,7 @@ strided_slice_d_op_info = TBERegOp("StridedSlice") \
.attr("shrink_axis_mask", "required", "int", "all") \
.input(0, "x", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/tensor_add.py
View File

@ -26,7 +26,7 @@ tensor_add_op_info = TBERegOp("Add") \
.input(0, "x1", False, "required", "all") \
.input(1, "x2", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.I32_Default, DataType.I32_Default, DataType.I32_Default) \
.dtype_format(DataType.F16_Default, DataType.F16_Default, DataType.F16_Default) \
.dtype_format(DataType.F32_Default, DataType.F32_Default, DataType.F32_Default) \

2
mindspore/ops/_op_impl/tbe/tensor_add_ds.py
View File

@ -27,7 +27,7 @@ tensor_add_op_info = TBERegOp("Add") \
.input(0, "x1", False, "required", "all") \
.input(1, "x2", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.I32_Default, DataType.I32_Default, DataType.I32_Default) \
.dtype_format(DataType.F16_Default, DataType.F16_Default, DataType.F16_Default) \
.dtype_format(DataType.F32_Default, DataType.F32_Default, DataType.F32_Default) \

2
mindspore/ops/_op_impl/tbe/tile.py
View File

@ -26,7 +26,7 @@ tile_op_info = TBERegOp("Tile") \
.attr("multiples", "optional", "listInt", "all")\
.input(0, "x1", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/unpack.py
View File

@ -27,7 +27,7 @@ unstack_op_info = TBERegOp("Unstack") \
.attr("axis", "required", "int", "all") \
.input(0, "x", False, "required", "all") \
.output(0, "y", False, "dynamic", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None) \
.get_op_info()

2
mindspore/ops/_op_impl/tbe/unsorted_segment_sum.py
View File

@ -27,7 +27,7 @@ unsorted_segment_sum_op_info = TBERegOp("UnsortedSegmentSum") \
.input(0, "x", False, "required", "all") \
.input(1, "segment_ids", False, "required", "all") \
.output(0, "y", False, "required", "all") \
.op_pattern("dynamicFormat") \
.is_dynamic_format(True) \
.dtype_format(DataType.None_None, DataType.None_None, DataType.None_None) \
.get_op_info()

26
mindspore/ops/op_info_register.py
View File

@ -353,9 +353,9 @@ class TBERegOp(RegOp):
self.kernel_name_ = ''
self.partial_flag_ = False
self.reshape_type_ = ''
self.dynamic_format_ = False
self.dynamic_shape_ = False
self.need_check_supported_ = False
self.is_dynamic_format_ = False
self.op_pattern_ = ""
def async_flag(self, async_flag):
@ -424,17 +424,6 @@ class TBERegOp(RegOp):
self.reshape_type_ = reshape_type
return self
def dynamic_format(self, dynamic_format):
"""
Whether the operator supports dynamic selection of format and dtype or not.
Args:
dynamic_format (bool): Value of dynamic format. Default: false.
"""
self._is_bool(dynamic_format)
self.dynamic_format_ = dynamic_format
return self
def dynamic_shape(self, dynamic_shape):
"""
Whether the operator supports dynamic shape.
@ -451,12 +440,23 @@ class TBERegOp(RegOp):
Whether the operator need check supports.
Args:
:param need_check_supported: (bool): Value of need_check_supported. Default: false.
need_check_supported (bool): Value of need_check_supported. Default: false.
"""
self._is_bool(need_check_supported)
self.need_check_supported_ = need_check_supported
return self
def is_dynamic_format(self, is_dynamic_format):
"""
Whether the operator need calop_select_format api.
Args:
is_dynamic_format (bool): Value of is_dynamic_format_. Default: false.
"""
self._is_bool(is_dynamic_format)
self.is_dynamic_format_ = is_dynamic_format
return self
def op_pattern(self, pattern=None):
"""
The behavior type of operator, such as broadcast, reduce and so on.