From 1c6ba29523880562ed3c9312975ee6d08aa94f70 Mon Sep 17 00:00:00 2001 From: taoxiangdong Date: Wed, 21 Oct 2020 16:26:23 +0800 Subject: [PATCH] resolve the system link is too long problem --- ge/client/proto/ge_api.proto | 1 + ge/client/proto/ge_ir.proto | 190 ++ ge/client/proto/insert_op.proto | 136 ++ ge/client/proto/om.proto | 396 ++++ ge/client/proto/task.proto | 165 ++ ge/common/proto/ge_ir.proto | 190 ++ ge/common/proto/insert_op.proto | 136 ++ ge/common/proto/om.proto | 396 ++++ ge/common/proto/op_mapping_info.proto | 73 + ge/common/proto/task.proto | 165 ++ ge/common/proto/tensorflow/attr_value.proto | 62 + ge/common/proto/tensorflow/function.proto | 100 + ge/common/proto/tensorflow/graph.proto | 56 + .../proto/tensorflow/graph_library.proto | 14 + ge/common/proto/tensorflow/node_def.proto | 63 + ge/common/proto/tensorflow/op_def.proto | 164 ++ .../proto/tensorflow/resource_handle.proto | 29 + ge/common/proto/tensorflow/tensor.proto | 94 + ge/common/proto/tensorflow/tensor_shape.proto | 45 + ge/common/proto/tensorflow/types.proto | 74 + ge/common/proto/tensorflow/versions.proto | 31 + ge/executor/proto/dump_task.proto | 111 + ge/executor/proto/ge_ir.proto | 190 ++ ge/executor/proto/insert_op.proto | 136 ++ ge/executor/proto/om.proto | 396 ++++ ge/executor/proto/op_mapping_info.proto | 73 + ge/executor/proto/task.proto | 165 ++ ge/offline/proto/ge_ir.proto | 190 ++ ge/offline/proto/insert_op.proto | 136 ++ ge/offline/proto/om.proto | 396 ++++ ge/offline/proto/task.proto | 165 ++ ge/proto/caffe/caffe.proto | 1821 +++++++++++++++++ ge/proto/dump_task.proto | 111 + ge/proto/fusion_model.proto | 21 + ge/proto/fwk_adapter.proto | 37 + ge/proto/ge_api.proto | 88 + ge/proto/ge_ir.proto | 190 ++ ge/proto/insert_op.proto | 136 ++ ge/proto/om.proto | 396 ++++ ge/proto/op_mapping_info.proto | 73 + ge/proto/optimizer_priority.proto | 7 + ge/proto/task.proto | 165 ++ ge/proto/tensorflow/attr_value.proto | 62 + ge/proto/tensorflow/function.proto | 100 + ge/proto/tensorflow/graph.proto | 56 + ge/proto/tensorflow/graph_library.proto | 14 + ge/proto/tensorflow/node_def.proto | 63 + ge/proto/tensorflow/op_def.proto | 164 ++ ge/proto/tensorflow/resource_handle.proto | 29 + ge/proto/tensorflow/tensor.proto | 94 + ge/proto/tensorflow/tensor_shape.proto | 45 + ge/proto/tensorflow/types.proto | 74 + ge/proto/tensorflow/versions.proto | 31 + 53 files changed, 8315 insertions(+) create mode 100644 ge/client/proto/ge_api.proto create mode 100644 ge/client/proto/ge_ir.proto create mode 100644 ge/client/proto/insert_op.proto create mode 100755 ge/client/proto/om.proto create mode 100644 ge/client/proto/task.proto create mode 100644 ge/common/proto/ge_ir.proto create mode 100644 ge/common/proto/insert_op.proto create mode 100644 ge/common/proto/om.proto create mode 100644 ge/common/proto/op_mapping_info.proto create mode 100644 ge/common/proto/task.proto create mode 100644 ge/common/proto/tensorflow/attr_value.proto create mode 100644 ge/common/proto/tensorflow/function.proto create mode 100644 ge/common/proto/tensorflow/graph.proto create mode 100644 ge/common/proto/tensorflow/graph_library.proto create mode 100644 ge/common/proto/tensorflow/node_def.proto create mode 100644 ge/common/proto/tensorflow/op_def.proto create mode 100644 ge/common/proto/tensorflow/resource_handle.proto create mode 100644 ge/common/proto/tensorflow/tensor.proto create mode 100644 ge/common/proto/tensorflow/tensor_shape.proto create mode 100644 ge/common/proto/tensorflow/types.proto create mode 100644 ge/common/proto/tensorflow/versions.proto create mode 100644 ge/executor/proto/dump_task.proto create mode 100644 ge/executor/proto/ge_ir.proto create mode 100644 ge/executor/proto/insert_op.proto create mode 100644 ge/executor/proto/om.proto create mode 100644 ge/executor/proto/op_mapping_info.proto create mode 100644 ge/executor/proto/task.proto create mode 100644 ge/offline/proto/ge_ir.proto create mode 100644 ge/offline/proto/insert_op.proto create mode 100644 ge/offline/proto/om.proto create mode 100644 ge/offline/proto/task.proto create mode 100644 ge/proto/caffe/caffe.proto create mode 100644 ge/proto/dump_task.proto create mode 100755 ge/proto/fusion_model.proto create mode 100644 ge/proto/fwk_adapter.proto create mode 100755 ge/proto/ge_api.proto create mode 100644 ge/proto/ge_ir.proto create mode 100644 ge/proto/insert_op.proto create mode 100644 ge/proto/om.proto create mode 100644 ge/proto/op_mapping_info.proto create mode 100644 ge/proto/optimizer_priority.proto create mode 100644 ge/proto/task.proto create mode 100644 ge/proto/tensorflow/attr_value.proto create mode 100644 ge/proto/tensorflow/function.proto create mode 100644 ge/proto/tensorflow/graph.proto create mode 100644 ge/proto/tensorflow/graph_library.proto create mode 100644 ge/proto/tensorflow/node_def.proto create mode 100644 ge/proto/tensorflow/op_def.proto create mode 100644 ge/proto/tensorflow/resource_handle.proto create mode 100644 ge/proto/tensorflow/tensor.proto create mode 100644 ge/proto/tensorflow/tensor_shape.proto create mode 100644 ge/proto/tensorflow/types.proto create mode 100644 ge/proto/tensorflow/versions.proto diff --git a/ge/client/proto/ge_api.proto b/ge/client/proto/ge_api.proto new file mode 100644 index 00000000..26d705fe --- /dev/null +++ b/ge/client/proto/ge_api.proto @@ -0,0 +1 @@ +../../proto/ge_api.proto \ No newline at end of file diff --git a/ge/client/proto/ge_ir.proto b/ge/client/proto/ge_ir.proto new file mode 100644 index 00000000..e7bfe0cb --- /dev/null +++ b/ge/client/proto/ge_ir.proto @@ -0,0 +1,190 @@ +syntax = "proto3"; + +package ge.proto; + +enum DataType +{ + DT_UNDEFINED = 0; // Used to indicate a DataType field has not been set. + DT_FLOAT = 1; // float type + DT_FLOAT16 = 2; // fp16 type + DT_INT8 = 3; // int8 type + DT_UINT8 = 4; // uint8 type + DT_INT16 = 5; // int16 type + DT_UINT16 = 6; // uint16 type + DT_INT32 = 7; // + DT_INT64 = 8; // int64 type + DT_UINT32 = 9; // unsigned int32 + DT_UINT64 = 10; // unsigned int64 + DT_BOOL = 11; // bool type + DT_DOUBLE = 12; // double type + DT_STRING = 13; // string type + DT_DUAL_SUB_INT8 = 14; /**< dual output int8 type */ + DT_DUAL_SUB_UINT8 = 15; /**< dual output uint8 type */ + DT_COMPLEX64 = 16; // complex64 type + DT_COMPLEX128 = 17; // complex128 type + DT_QINT8 = 18; // qint8 type + DT_QINT16 = 19; // qint16 type + DT_QINT32 = 20; // qint32 type + DT_QUINT8 = 21; // quint8 type + DT_QUINT16 = 22; // quint16 type + DT_RESOURCE = 23; // resource type + DT_STRING_REF = 24; // string_ref type + DT_DUAL = 25; /**< dual output type */ +} + +message AttrDef +{ + message ListValue + { + enum ListValueType{ + VT_LIST_NONE = 0; + VT_LIST_STRING = 1; + VT_LIST_INT = 2; + VT_LIST_FLOAT = 3; + VT_LIST_BOOL = 4; + VT_LIST_BYTES = 5; + VT_LIST_TENSOR_DESC = 6; + VT_LIST_TENSOR = 7; + VT_LIST_GRAPH = 8; + VT_LIST_NAMED_ATTRS = 9; + VT_LIST_DATA_TYPE = 10; + } + repeated bytes s = 2; // "list(string)" + repeated int64 i = 3; // "list(int)" + repeated float f = 4; // "list(float)" + repeated bool b = 5; // "list(bool)" + repeated bytes bt = 7; + repeated TensorDescriptor td = 8; + repeated TensorDef t = 9; + repeated GraphDef g = 10; + repeated NamedAttrs na = 11; + repeated int64 dt = 12; // list ge::DataType + + ListValueType val_type = 20; + } + + message ListListInt{ + message ListInt{ + repeated int64 list_i = 1; // list int + } + repeated ListInt list_list_i = 1; // list list int + } + + oneof value + { + bytes s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + bytes bt = 7; + ListValue list = 1; // any "list(...)" + NamedAttrs func = 10; // Used to support attr nesting + TensorDescriptor td = 11; // GeTensorDesc type + TensorDef t = 12; // GeTensor type + GraphDef g = 13; // Graph type + ListListInt list_list_int = 14; // List List Int type + int64 dt = 15; // ge::DataType + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NamedAttrs +{ + string name = 1; + map attr = 2; +} + +// Shape / dimension description, using row-major order +message ShapeDef +{ + repeated int64 dim = 1; // Size of each dimension +} + +// Multidimensional data description +message TensorDescriptor +{ + string name = 1; // Optional parameter, tensor name + + DataType dtype = 2; // tensor datatype + ShapeDef shape = 3; // Shape / dimension + string layout = 4; // Tensor format, eg: "NCHW", "NHWC", "CHW", "ND" + + bool has_out_attr = 9; + int64 size = 10; + int64 weight_size = 11; + bool reuse_input = 12; + bool output_tensor = 13; + string device_type = 14; + bool input_tensor =15; + int64 real_dim_cnt = 16; + int64 reuse_input_index = 17; + int64 data_offset = 18; + int64 cmps_size = 19; + string cmps_tab = 20; + int64 cmps_tab_offset = 21; + + map attr = 5; // Set of extra parameter fields +} + +// GeTensor definition +message TensorDef +{ + TensorDescriptor desc = 1; // Tensor description + bytes data = 2; // Tensor data +} + + +// Operator description +message OpDef +{ + string name = 1; // name + string type = 2; // type + + repeated string input = 5; // input original op name + outgoing index. op_name锛歩ndex + + map attr = 10; // Set of operator parameter fields + + bool has_out_attr = 20; + int64 id = 21; + int64 stream_id =22; + repeated string input_name = 23; + repeated string src_name = 24; + repeated int64 src_index = 25; + repeated string dst_name = 26; + repeated int64 dst_index = 27; + repeated int64 input_i = 28; + repeated int64 output_i = 29; + repeated int64 workspace = 30; + repeated int64 workspace_bytes = 31; + repeated bool is_input_const = 32; + repeated TensorDescriptor input_desc = 33; + repeated TensorDescriptor output_desc = 34; + repeated string subgraph_name = 35; +} + +// Graph definition +message GraphDef +{ + string name = 1; // name + + repeated string input = 4; // Graph input + repeated string output = 5; // Graph output + + repeated OpDef op = 6; // List of operators + + map attr = 11; // Extended field +} + +// model definition +message ModelDef +{ + string name = 1; // name + uint32 version = 2; // IR Proto verion + string custom_version = 3; // User model version number, passed in by user + + repeated GraphDef graph = 7; // Graph definition锛実raph[0] represents the main diagram in modeldef + + map attr = 11; // Extended field +} + diff --git a/ge/client/proto/insert_op.proto b/ge/client/proto/insert_op.proto new file mode 100644 index 00000000..c635ca14 --- /dev/null +++ b/ge/client/proto/insert_op.proto @@ -0,0 +1,136 @@ +syntax = "proto3"; + +package domi; + +message InsertNewOps { + repeated AippOpParams aipp_op = 1; + repeated MultiShapeOpParams multi_shape_op = 2; +} + +message AippOpParams { + enum InputFormat { + UNDEFINED = 0; + YUV420SP_U8 = 1; + XRGB8888_U8 = 2; + RGB888_U8 = 3; + YUV400_U8 = 4; + NC1HWC0DI_FP16 = 5; + NC1HWC0DI_S8 = 6; + ARGB8888_U8 = 7; + YUYV_U8 = 8; + YUV422SP_U8 = 9; + AYUV444_U8 = 10; + RAW10 = 11; + RAW12 = 12; + RAW16 = 13; + RAW24 = 14; + RGB16 = 15; + RGB20 = 16; + RGB24 = 17; + RGB8_IR = 18; + RGB16_IR = 19; + RGB24_IR = 20; + } + + enum AippMode { + undefined = 0; + static = 1; + dynamic = 2; + } + + // AIPP模式，区分静态AIPP和动态AIPP + AippMode aipp_mode = 1; + + // related_input_rank参数为必填，类型为整型，配置范围>=0, <=输入Data算子的个数，默认值为0。 + // 标识对模型的第几个输入做AIPP处理，例如模型有两个输入，需要对第2个输入做AIPP，则配置related_input_rank为1。 + uint32 related_input_rank = 2; + + // input_edge_idx参数为可选，类型为整型，配置范围为>=0。 + // 配置该参数的作用，在于对Data算子不同的输出做不同的AIPP处理，如果该参数没有配置，默认对related_input_rank指定的模型输入的所有输出边做AIPP。 + // 配置值 <= Data算子输出边的个数。 + repeated uint32 input_edge_idx = 3; + + // [Begin] 动态AIPP参数，配置静态AIPP时无效 + uint32 max_src_image_size = 4; + + // 是否支持旋转。默认不支持，开启支持旋转时，会有额外的空间和性能损失 + bool support_rotation = 5; + + // [End] 动态AIPP参数 + + + // [Begin] 静态AIPP参数，配置动态AIPP时无效 + InputFormat input_format = 51; + bool csc_switch = 52; + float cpadding_value = 53; + bool rbuv_swap_switch = 54; + bool ax_swap_switch = 55; + bool single_line_mode = 56; + + int32 src_image_size_w = 57; + int32 src_image_size_h = 58; + + bool crop = 59; + int32 load_start_pos_w = 60; + int32 load_start_pos_h = 61; + int32 crop_size_w = 62; + int32 crop_size_h = 63; + + bool resize = 64; + int32 resize_output_w = 65; + int32 resize_output_h = 66; + + bool padding = 67; + int32 left_padding_size = 68; + int32 right_padding_size = 69; + int32 top_padding_size = 70; + int32 bottom_padding_size = 71; + + int32 mean_chn_0 = 10; + int32 mean_chn_1 = 11; + int32 mean_chn_2 = 12; + int32 mean_chn_3 = 19; + float min_chn_0 = 13; + float min_chn_1 = 14; + float min_chn_2 = 15; + float min_chn_3 = 20; + repeated float var_reci_chn_0 = 16; + repeated float var_reci_chn_1 = 17; + repeated float var_reci_chn_2 = 18; + repeated float var_reci_chn_3 = 21; + + repeated int32 matrix_r0c0 = 30; + repeated int32 matrix_r0c1 = 31; + repeated int32 matrix_r0c2 = 32; + repeated int32 matrix_r1c0 = 33; + repeated int32 matrix_r1c1 = 34; + repeated int32 matrix_r1c2 = 35; + repeated int32 matrix_r2c0 = 36; + repeated int32 matrix_r2c1 = 37; + repeated int32 matrix_r2c2 = 38; + repeated int32 output_bias_0 = 39; + repeated int32 output_bias_1 = 40; + repeated int32 output_bias_2 = 41; + repeated int32 input_bias_0 = 42; + repeated int32 input_bias_1 = 43; + repeated int32 input_bias_2 = 44; + + // [End] 静态AIPP参数 + + // The n number that is used for raw/rgbir data into f16 transformation. + // The transformation equation is x/(2^n). If set to 0, no transform is performed. + uint32 raw_rgbir_to_f16_n = 45; +} + +message MultiShapeOpParams { + enum MultiShapeMode { + batch = 0; //动态batch + resolution = 1; //动态分辨率，扩展用 + } + + MultiShapeMode mode = 1; //算子模式 + uint32 related_input_rank = 2; //新增算子插入到哪个输入 + + + repeated uint32 batch_list = 11; //batch_list值，batch_list的个数是2到8之间 +} diff --git a/ge/client/proto/om.proto b/ge/client/proto/om.proto new file mode 100755 index 00000000..e15e5f80 --- /dev/null +++ b/ge/client/proto/om.proto @@ -0,0 +1,396 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +package domi; + +enum TargetType +{ + MINI = 0; + TINY = 1; + LITE = 2; +} + +// offline model +message ModelDef { + string name = 1; + uint32 version = 2; + + uint64 memory_size = 10; + uint32 stream_num = 11; + uint32 event_num = 12; + uint64 weight_size = 13; + uint32 label_num = 15; + repeated OpDef op = 20; + TargetType target_type = 23; + + map attr = 30; +}; + +// operator define +message OpDef { + string name = 1; + string type = 2; + + uint32 id = 3; + uint32 stream_id = 4; + + repeated string input_name = 5; + + repeated string src_name = 8; + repeated int32 src_index = 9; + repeated int64 input = 10; + repeated int64 output = 11; + repeated TensorDescriptor input_desc = 12; + repeated TensorDescriptor output_desc = 13; + repeated WeightDef weights = 14; + repeated string dst_name = 15; + repeated int32 dst_index = 16; + + repeated int64 workspace = 20; + repeated uint32 workspace_bytes = 21; + + repeated string weight_name = 22; + repeated bool is_input_const = 23; + + map attr = 30; + + QuantizeFactorParams quantize_factor = 31; + + oneof op_params { + // start at 100 here + SendOpParams sender_param = 100; + RecvOpParams receiver_param = 200; + ConvolutionOpParams convolution_param = 300; + PoolingOpParams pooling_param = 400; + EltwiseOpParams eltwise_param = 500; + BatchNormOpParams batchnorm_param = 600; + ScaleOpParams scale_param = 700; + FullConnectionOpParams full_connection_param = 800; + SoftmaxOpParams softmax_param = 900; + ActivationOpParams activation_param = 1000; + ReshapeOpParams reshape_param = 1100; + } +}; + +message SendOpParams { + uint32 event_id = 1; +}; + +message RecvOpParams { + uint32 event_id = 1; +}; + +enum QuantizeScaleType +{ + VECTOR_SCALE = 0; + SCALAR_SCALE = 1; +} + +enum QuantizeScaleMode +{ + NORMAL_MODE = 0; + SQRT_MODE = 1; +} + +enum QuantizeAlgorithm +{ + NON_OFFSET_ALGO = 0; + HALF_OFFSET_ALGO = 1; + ALL_OFFSET_ALGO = 2; +} +message QuantizeFactor +{ + QuantizeScaleMode scale_mode = 1; + bytes scale_value = 2; + int64 scale_offset = 3; + bytes offset_data_value = 4; + int64 offset_data_offset = 5; + bytes offset_weight_value = 6; + int64 offset_weight_offset = 7; + bytes offset_pad_value = 8; + int64 offset_pad_offset = 9; +}; + +message QuantizeCalcFactor +{ + bytes offsetw = 1; + int64 offsetw_offset = 2; + bytes offsetd = 3; + int64 offsetd_offset = 4; + bytes scalereq = 5; + int64 scaledreq_offset = 6; + bytes offsetdnext = 7; + int64 offsetdnext_offset = 8; +} + +message QuantizeFactorParams +{ + QuantizeAlgorithm quantize_algo = 1; + QuantizeScaleType scale_type = 2; + QuantizeFactor quantize_param = 3; + QuantizeFactor dequantize_param = 4; + QuantizeFactor requantize_param = 5; + QuantizeCalcFactor quantizecalc_param = 6; +}; + +message ConvolutionOpParams { + int32 mode = 1; + int32 algo = 2; + int32 pad_mode = 3; + uint32 group = 4; + uint32 num_output = 5; + + repeated uint32 pad = 10; + repeated uint32 stride = 11; + repeated uint32 dilation = 12; + repeated uint32 kernel = 13; + + float alpha = 20; + float beta = 21; + + WeightDef filter = 40; + WeightDef bias = 41; + + bool relu_flag = 62; + repeated uint32 adj = 70; + repeated uint32 target_shape = 71; + repeated uint32 before_pad = 72; +}; + +message PoolingOpParams { + int32 mode = 1; + int32 nan_opt = 2; + int32 pad_mode = 3; + bool global_pooling = 4; + + repeated uint32 window = 10; + repeated uint32 pad = 11; + repeated uint32 stride = 12; + bool ceil_mode = 13; + int32 data_mode = 14; + + float alpha = 20; + float beta = 21; + repeated uint32 before_pad = 22; +}; + +message EltwiseOpParams { + int32 mode = 1; + repeated float coeff = 2; + float alpha = 3; + float beta = 4; + repeated WeightDef weight = 5; + bool relu_flag = 6; +}; + +message ActivationOpParams { + int32 mode = 1; + float coef = 2; + float alpha = 3; + float beta = 4; +}; + +message BatchNormOpParams { + int32 mode = 1; + + float alpha = 2; + float beta = 3; + double epsilon = 4;//optinal,[default = 1e-5] + bool use_global_stats = 5; //optinal,by default true,testing mode + float moving_average_fraction = 6; //optinal,[default = .999]; + + WeightDef estimated_mean = 7; + WeightDef estimated_variance = 8; + + WeightDef scale = 9; + WeightDef bias = 10; +}; + +message ScaleOpParams { + WeightDef scale = 1; + WeightDef bias = 2; +}; + +message ReshapeOpParams { + float alpha = 1; + float beta = 2; + ShapeDef shape = 3; + int32 axis = 4; + int32 num_axes = 5; + int32 format = 6; +}; + +message SoftmaxOpParams { + int32 algo = 1; + int32 mode = 2; + float alpha = 3; + float beta = 4; +}; + +message FullConnectionOpParams { + WeightDef filter = 1; + WeightDef bias = 2; + uint32 num_output = 3; + bool relu_flag = 12; +}; + +message FlattenOpParams { + float alpha = 1; + float beta = 2; + int32 start_axis = 3; + int32 end_axis = 4; +} + +message AddLimitedOpParams { + float alpha = 1; + float beta = 2; + int32 axis = 3; + bool broadcast = 4; + + repeated WeightDef weight = 10; +}; + +message MulLimitedOpParams { + float alpha = 1; + float beta = 2; + int32 axis = 3; + bool broadcast = 4; + + repeated WeightDef weight = 10; +}; + +message AddOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message MulOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message SubOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message BiasAddOpParams { + float alpha = 1; + float beta = 2; + + WeightDef bias = 10; +}; + +message MatMulOpParams { + float alpha = 1; + float beta = 2; + bool transposeX = 3; + bool transposeW = 4; + + WeightDef filter = 10; + WeightDef bias = 12; +}; + +message RsqrtOpParams { + float alpha = 1; + float beta = 2; +}; + + +message WeightDef { + int32 format = 1; + int32 data_type = 2; + ShapeDef shape = 3; + bytes data = 4; + int64 data_offset = 5; + uint32 cmps_size = 6; + bytes cmps_tab = 7; + int64 cmps_tab_offset = 10; + CompressInfo cmps_info = 8; + AllOffsetQuantizeInfo alloffset_quantize_info = 11; +} + +message ShapeDef { + repeated int64 dim = 1; +} + +enum DeviceType { + NPU = 0; // In default, we will use NPU. + CPU = 1; // CPU +} + +message AllOffsetQuantizeInfo { + float scale = 1; + int32 offset = 2; +} + +message TensorDescriptor { + int32 format = 1; + int32 data_type = 2; + repeated int64 dim = 3; + uint32 size = 4; + bool reuse_input = 5; + bool output_tensor = 7; + DeviceType device_type = 8; + bool input_tensor = 9; + uint32 real_dim_cnt = 10; + uint32 reuse_input_index = 11; + AllOffsetQuantizeInfo alloffset_quantize_info = 12; +} + +message CompressInfo { + int32 blockRow = 1; // block row + int32 blockCol = 2; // block col + int32 fractalK = 3; // fractal K + int32 fractalN = 4; // fractal N + int32 lastFractalK = 5; // K of last fractal + int32 lastFractalN = 6; // N of last fractal + int32 cubeSize = 7; // cube's length + int32 loadDir = 8; // data load directtiono 0:col load 1:row load +} + +message AttrDef { + message ListValue { + repeated string s = 2; // "list(string)" + repeated int64 i = 3 [packed = true]; // "list(int)" + repeated float f = 4 [packed = true]; // "list(float)" + repeated bool b = 5 [packed = true]; // "list(bool)" + repeated uint32 u = 6 [packed = true]; // "list(uint)" + repeated bytes bt = 7; + } + + oneof value { + string s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + uint32 u = 6; // "uint32" + bytes bt = 7; + ListValue list = 1; // any "list(...)" + NamedAttrs func = 10; + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NamedAttrs { + string name = 1; + map attr = 2; +} + diff --git a/ge/client/proto/task.proto b/ge/client/proto/task.proto new file mode 100644 index 00000000..d0c09840 --- /dev/null +++ b/ge/client/proto/task.proto @@ -0,0 +1,165 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +package domi; + +message ModelTaskDef { + string version = 1; + + map attr = 9; // Extended field + repeated TaskDef task = 10; + + uint64 memory_size = 11; + uint32 stream_num = 12; + uint32 event_num = 13; + uint64 weight_size = 14; + + repeated bytes op = 15; // input/output opdef in bytes + + uint64 base_addr = 16; // base addr + uint64 weight_addr = 17; // weight addr + uint32 batch_num = 18; +} + + +message TaskDef { + uint32 id = 1; + uint32 type = 2; + + uint32 stream_id = 10; + uint32 event_id = 11; + + KernelDef kernel = 20; + KernelExDef kernel_ex = 21; + KernelHcclDef kernel_hccl = 25; + EventExDef event_ex = 26; + LogTimeStampDef log_timestamp = 28; + + uint32 label_id = 30; + + MemcpyAsyncDef memcpy_async = 31; + StreamSwitchDef stream_switch = 32; + StreamActiveDef stream_active = 33; + bytes private_def = 34; + uint64 ops_kernel_store_ptr = 35; // adjustments to other fields in the future + StreamSwitchNDef stream_switch_n = 36; + + LabelSetDef label_set = 37; + LabelGotoExDef label_goto_ex = 38; + LabelSwitchByIndexDef label_switch_by_index = 39; +} + +message KernelDef { + KernelContext context = 1; + + string stub_func = 10; + uint32 block_dim = 11; + uint32 args_size = 12; + bytes args = 13; + bytes sm_desc = 14; + bytes flowtable = 15; + string so_name = 16; + string kernel_name = 17; + bytes kernel_ext_info = 18; + uint32 kernel_ext_info_size = 19; +} + +message KernelContext { + uint32 kernel_type = 1; + uint32 op_id = 2; // OP type in CCE + uint32 kernel_func_id = 3; + uint32 op_index = 4; // TE/Custom operator + bool is_flowtable = 5; // Identify whether args is a flowtable structure + bytes args_offset = 6; // args offset information + uint32 args_count = 7; // args count + repeated uint32 origin_op_index = 8; +} + + +message KernelExDef { + uint32 flags = 1; + + uint32 op_index = 4; + uint32 args_size = 12; + bytes args = 13; + bytes task_info = 14; // serialized nodeDef, funcDef, inputoutput + uint32 task_info_size = 15; + bytes kernel_ext_info = 16; + uint32 kernel_ext_info_size = 17; +} + + +message KernelHcclDef { + uint32 op_index = 8; + string hccl_type = 9; +} + + +message EventExDef { + uint32 op_index = 1; + uint32 event_type = 2; +} + +message LogTimeStampDef { + uint64 logid = 1; + bool notify = 2; + uint32 flat = 3; +} + +message MemcpyAsyncDef { + uint64 dst = 1; + uint64 dst_max = 2; + uint64 src = 3; + uint64 count = 4; + uint32 kind = 5; + uint32 op_index = 6; +} + +message StreamSwitchDef { + uint32 op_index = 1; + uint32 true_stream_id = 2; + int64 value = 3; + uint64 value_ptr = 4; + uint32 data_type = 5; +} + +message StreamActiveDef { + uint32 op_index = 1; + uint32 active_stream_id = 2; +} + +message StreamSwitchNDef { + uint32 op_index = 1; + uint32 size = 2; + repeated int64 target_value = 3; + repeated uint32 true_stream_id = 4; + uint32 element_size = 5; + uint32 data_type = 6; +} + +message LabelSetDef { + uint32 op_index = 1; + uint32 label_id = 2; + uint32 model_id = 3; +} + +message LabelGotoExDef { + uint32 op_index = 1; + uint32 label_id = 2; + uint32 model_id = 3; +} + +message LabelSwitchByIndexDef { + uint32 op_index = 1; + uint32 label_max = 2; +} diff --git a/ge/common/proto/ge_ir.proto b/ge/common/proto/ge_ir.proto new file mode 100644 index 00000000..e7bfe0cb --- /dev/null +++ b/ge/common/proto/ge_ir.proto @@ -0,0 +1,190 @@ +syntax = "proto3"; + +package ge.proto; + +enum DataType +{ + DT_UNDEFINED = 0; // Used to indicate a DataType field has not been set. + DT_FLOAT = 1; // float type + DT_FLOAT16 = 2; // fp16 type + DT_INT8 = 3; // int8 type + DT_UINT8 = 4; // uint8 type + DT_INT16 = 5; // int16 type + DT_UINT16 = 6; // uint16 type + DT_INT32 = 7; // + DT_INT64 = 8; // int64 type + DT_UINT32 = 9; // unsigned int32 + DT_UINT64 = 10; // unsigned int64 + DT_BOOL = 11; // bool type + DT_DOUBLE = 12; // double type + DT_STRING = 13; // string type + DT_DUAL_SUB_INT8 = 14; /**< dual output int8 type */ + DT_DUAL_SUB_UINT8 = 15; /**< dual output uint8 type */ + DT_COMPLEX64 = 16; // complex64 type + DT_COMPLEX128 = 17; // complex128 type + DT_QINT8 = 18; // qint8 type + DT_QINT16 = 19; // qint16 type + DT_QINT32 = 20; // qint32 type + DT_QUINT8 = 21; // quint8 type + DT_QUINT16 = 22; // quint16 type + DT_RESOURCE = 23; // resource type + DT_STRING_REF = 24; // string_ref type + DT_DUAL = 25; /**< dual output type */ +} + +message AttrDef +{ + message ListValue + { + enum ListValueType{ + VT_LIST_NONE = 0; + VT_LIST_STRING = 1; + VT_LIST_INT = 2; + VT_LIST_FLOAT = 3; + VT_LIST_BOOL = 4; + VT_LIST_BYTES = 5; + VT_LIST_TENSOR_DESC = 6; + VT_LIST_TENSOR = 7; + VT_LIST_GRAPH = 8; + VT_LIST_NAMED_ATTRS = 9; + VT_LIST_DATA_TYPE = 10; + } + repeated bytes s = 2; // "list(string)" + repeated int64 i = 3; // "list(int)" + repeated float f = 4; // "list(float)" + repeated bool b = 5; // "list(bool)" + repeated bytes bt = 7; + repeated TensorDescriptor td = 8; + repeated TensorDef t = 9; + repeated GraphDef g = 10; + repeated NamedAttrs na = 11; + repeated int64 dt = 12; // list ge::DataType + + ListValueType val_type = 20; + } + + message ListListInt{ + message ListInt{ + repeated int64 list_i = 1; // list int + } + repeated ListInt list_list_i = 1; // list list int + } + + oneof value + { + bytes s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + bytes bt = 7; + ListValue list = 1; // any "list(...)" + NamedAttrs func = 10; // Used to support attr nesting + TensorDescriptor td = 11; // GeTensorDesc type + TensorDef t = 12; // GeTensor type + GraphDef g = 13; // Graph type + ListListInt list_list_int = 14; // List List Int type + int64 dt = 15; // ge::DataType + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NamedAttrs +{ + string name = 1; + map attr = 2; +} + +// Shape / dimension description, using row-major order +message ShapeDef +{ + repeated int64 dim = 1; // Size of each dimension +} + +// Multidimensional data description +message TensorDescriptor +{ + string name = 1; // Optional parameter, tensor name + + DataType dtype = 2; // tensor datatype + ShapeDef shape = 3; // Shape / dimension + string layout = 4; // Tensor format, eg: "NCHW", "NHWC", "CHW", "ND" + + bool has_out_attr = 9; + int64 size = 10; + int64 weight_size = 11; + bool reuse_input = 12; + bool output_tensor = 13; + string device_type = 14; + bool input_tensor =15; + int64 real_dim_cnt = 16; + int64 reuse_input_index = 17; + int64 data_offset = 18; + int64 cmps_size = 19; + string cmps_tab = 20; + int64 cmps_tab_offset = 21; + + map attr = 5; // Set of extra parameter fields +} + +// GeTensor definition +message TensorDef +{ + TensorDescriptor desc = 1; // Tensor description + bytes data = 2; // Tensor data +} + + +// Operator description +message OpDef +{ + string name = 1; // name + string type = 2; // type + + repeated string input = 5; // input original op name + outgoing index. op_name锛歩ndex + + map attr = 10; // Set of operator parameter fields + + bool has_out_attr = 20; + int64 id = 21; + int64 stream_id =22; + repeated string input_name = 23; + repeated string src_name = 24; + repeated int64 src_index = 25; + repeated string dst_name = 26; + repeated int64 dst_index = 27; + repeated int64 input_i = 28; + repeated int64 output_i = 29; + repeated int64 workspace = 30; + repeated int64 workspace_bytes = 31; + repeated bool is_input_const = 32; + repeated TensorDescriptor input_desc = 33; + repeated TensorDescriptor output_desc = 34; + repeated string subgraph_name = 35; +} + +// Graph definition +message GraphDef +{ + string name = 1; // name + + repeated string input = 4; // Graph input + repeated string output = 5; // Graph output + + repeated OpDef op = 6; // List of operators + + map attr = 11; // Extended field +} + +// model definition +message ModelDef +{ + string name = 1; // name + uint32 version = 2; // IR Proto verion + string custom_version = 3; // User model version number, passed in by user + + repeated GraphDef graph = 7; // Graph definition锛実raph[0] represents the main diagram in modeldef + + map attr = 11; // Extended field +} + diff --git a/ge/common/proto/insert_op.proto b/ge/common/proto/insert_op.proto new file mode 100644 index 00000000..c635ca14 --- /dev/null +++ b/ge/common/proto/insert_op.proto @@ -0,0 +1,136 @@ +syntax = "proto3"; + +package domi; + +message InsertNewOps { + repeated AippOpParams aipp_op = 1; + repeated MultiShapeOpParams multi_shape_op = 2; +} + +message AippOpParams { + enum InputFormat { + UNDEFINED = 0; + YUV420SP_U8 = 1; + XRGB8888_U8 = 2; + RGB888_U8 = 3; + YUV400_U8 = 4; + NC1HWC0DI_FP16 = 5; + NC1HWC0DI_S8 = 6; + ARGB8888_U8 = 7; + YUYV_U8 = 8; + YUV422SP_U8 = 9; + AYUV444_U8 = 10; + RAW10 = 11; + RAW12 = 12; + RAW16 = 13; + RAW24 = 14; + RGB16 = 15; + RGB20 = 16; + RGB24 = 17; + RGB8_IR = 18; + RGB16_IR = 19; + RGB24_IR = 20; + } + + enum AippMode { + undefined = 0; + static = 1; + dynamic = 2; + } + + // AIPP模式，区分静态AIPP和动态AIPP + AippMode aipp_mode = 1; + + // related_input_rank参数为必填，类型为整型，配置范围>=0, <=输入Data算子的个数，默认值为0。 + // 标识对模型的第几个输入做AIPP处理，例如模型有两个输入，需要对第2个输入做AIPP，则配置related_input_rank为1。 + uint32 related_input_rank = 2; + + // input_edge_idx参数为可选，类型为整型，配置范围为>=0。 + // 配置该参数的作用，在于对Data算子不同的输出做不同的AIPP处理，如果该参数没有配置，默认对related_input_rank指定的模型输入的所有输出边做AIPP。 + // 配置值 <= Data算子输出边的个数。 + repeated uint32 input_edge_idx = 3; + + // [Begin] 动态AIPP参数，配置静态AIPP时无效 + uint32 max_src_image_size = 4; + + // 是否支持旋转。默认不支持，开启支持旋转时，会有额外的空间和性能损失 + bool support_rotation = 5; + + // [End] 动态AIPP参数 + + + // [Begin] 静态AIPP参数，配置动态AIPP时无效 + InputFormat input_format = 51; + bool csc_switch = 52; + float cpadding_value = 53; + bool rbuv_swap_switch = 54; + bool ax_swap_switch = 55; + bool single_line_mode = 56; + + int32 src_image_size_w = 57; + int32 src_image_size_h = 58; + + bool crop = 59; + int32 load_start_pos_w = 60; + int32 load_start_pos_h = 61; + int32 crop_size_w = 62; + int32 crop_size_h = 63; + + bool resize = 64; + int32 resize_output_w = 65; + int32 resize_output_h = 66; + + bool padding = 67; + int32 left_padding_size = 68; + int32 right_padding_size = 69; + int32 top_padding_size = 70; + int32 bottom_padding_size = 71; + + int32 mean_chn_0 = 10; + int32 mean_chn_1 = 11; + int32 mean_chn_2 = 12; + int32 mean_chn_3 = 19; + float min_chn_0 = 13; + float min_chn_1 = 14; + float min_chn_2 = 15; + float min_chn_3 = 20; + repeated float var_reci_chn_0 = 16; + repeated float var_reci_chn_1 = 17; + repeated float var_reci_chn_2 = 18; + repeated float var_reci_chn_3 = 21; + + repeated int32 matrix_r0c0 = 30; + repeated int32 matrix_r0c1 = 31; + repeated int32 matrix_r0c2 = 32; + repeated int32 matrix_r1c0 = 33; + repeated int32 matrix_r1c1 = 34; + repeated int32 matrix_r1c2 = 35; + repeated int32 matrix_r2c0 = 36; + repeated int32 matrix_r2c1 = 37; + repeated int32 matrix_r2c2 = 38; + repeated int32 output_bias_0 = 39; + repeated int32 output_bias_1 = 40; + repeated int32 output_bias_2 = 41; + repeated int32 input_bias_0 = 42; + repeated int32 input_bias_1 = 43; + repeated int32 input_bias_2 = 44; + + // [End] 静态AIPP参数 + + // The n number that is used for raw/rgbir data into f16 transformation. + // The transformation equation is x/(2^n). If set to 0, no transform is performed. + uint32 raw_rgbir_to_f16_n = 45; +} + +message MultiShapeOpParams { + enum MultiShapeMode { + batch = 0; //动态batch + resolution = 1; //动态分辨率，扩展用 + } + + MultiShapeMode mode = 1; //算子模式 + uint32 related_input_rank = 2; //新增算子插入到哪个输入 + + + repeated uint32 batch_list = 11; //batch_list值，batch_list的个数是2到8之间 +} diff --git a/ge/common/proto/om.proto b/ge/common/proto/om.proto new file mode 100644 index 00000000..e15e5f80 --- /dev/null +++ b/ge/common/proto/om.proto @@ -0,0 +1,396 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +package domi; + +enum TargetType +{ + MINI = 0; + TINY = 1; + LITE = 2; +} + +// offline model +message ModelDef { + string name = 1; + uint32 version = 2; + + uint64 memory_size = 10; + uint32 stream_num = 11; + uint32 event_num = 12; + uint64 weight_size = 13; + uint32 label_num = 15; + repeated OpDef op = 20; + TargetType target_type = 23; + + map attr = 30; +}; + +// operator define +message OpDef { + string name = 1; + string type = 2; + + uint32 id = 3; + uint32 stream_id = 4; + + repeated string input_name = 5; + + repeated string src_name = 8; + repeated int32 src_index = 9; + repeated int64 input = 10; + repeated int64 output = 11; + repeated TensorDescriptor input_desc = 12; + repeated TensorDescriptor output_desc = 13; + repeated WeightDef weights = 14; + repeated string dst_name = 15; + repeated int32 dst_index = 16; + + repeated int64 workspace = 20; + repeated uint32 workspace_bytes = 21; + + repeated string weight_name = 22; + repeated bool is_input_const = 23; + + map attr = 30; + + QuantizeFactorParams quantize_factor = 31; + + oneof op_params { + // start at 100 here + SendOpParams sender_param = 100; + RecvOpParams receiver_param = 200; + ConvolutionOpParams convolution_param = 300; + PoolingOpParams pooling_param = 400; + EltwiseOpParams eltwise_param = 500; + BatchNormOpParams batchnorm_param = 600; + ScaleOpParams scale_param = 700; + FullConnectionOpParams full_connection_param = 800; + SoftmaxOpParams softmax_param = 900; + ActivationOpParams activation_param = 1000; + ReshapeOpParams reshape_param = 1100; + } +}; + +message SendOpParams { + uint32 event_id = 1; +}; + +message RecvOpParams { + uint32 event_id = 1; +}; + +enum QuantizeScaleType +{ + VECTOR_SCALE = 0; + SCALAR_SCALE = 1; +} + +enum QuantizeScaleMode +{ + NORMAL_MODE = 0; + SQRT_MODE = 1; +} + +enum QuantizeAlgorithm +{ + NON_OFFSET_ALGO = 0; + HALF_OFFSET_ALGO = 1; + ALL_OFFSET_ALGO = 2; +} +message QuantizeFactor +{ + QuantizeScaleMode scale_mode = 1; + bytes scale_value = 2; + int64 scale_offset = 3; + bytes offset_data_value = 4; + int64 offset_data_offset = 5; + bytes offset_weight_value = 6; + int64 offset_weight_offset = 7; + bytes offset_pad_value = 8; + int64 offset_pad_offset = 9; +}; + +message QuantizeCalcFactor +{ + bytes offsetw = 1; + int64 offsetw_offset = 2; + bytes offsetd = 3; + int64 offsetd_offset = 4; + bytes scalereq = 5; + int64 scaledreq_offset = 6; + bytes offsetdnext = 7; + int64 offsetdnext_offset = 8; +} + +message QuantizeFactorParams +{ + QuantizeAlgorithm quantize_algo = 1; + QuantizeScaleType scale_type = 2; + QuantizeFactor quantize_param = 3; + QuantizeFactor dequantize_param = 4; + QuantizeFactor requantize_param = 5; + QuantizeCalcFactor quantizecalc_param = 6; +}; + +message ConvolutionOpParams { + int32 mode = 1; + int32 algo = 2; + int32 pad_mode = 3; + uint32 group = 4; + uint32 num_output = 5; + + repeated uint32 pad = 10; + repeated uint32 stride = 11; + repeated uint32 dilation = 12; + repeated uint32 kernel = 13; + + float alpha = 20; + float beta = 21; + + WeightDef filter = 40; + WeightDef bias = 41; + + bool relu_flag = 62; + repeated uint32 adj = 70; + repeated uint32 target_shape = 71; + repeated uint32 before_pad = 72; +}; + +message PoolingOpParams { + int32 mode = 1; + int32 nan_opt = 2; + int32 pad_mode = 3; + bool global_pooling = 4; + + repeated uint32 window = 10; + repeated uint32 pad = 11; + repeated uint32 stride = 12; + bool ceil_mode = 13; + int32 data_mode = 14; + + float alpha = 20; + float beta = 21; + repeated uint32 before_pad = 22; +}; + +message EltwiseOpParams { + int32 mode = 1; + repeated float coeff = 2; + float alpha = 3; + float beta = 4; + repeated WeightDef weight = 5; + bool relu_flag = 6; +}; + +message ActivationOpParams { + int32 mode = 1; + float coef = 2; + float alpha = 3; + float beta = 4; +}; + +message BatchNormOpParams { + int32 mode = 1; + + float alpha = 2; + float beta = 3; + double epsilon = 4;//optinal,[default = 1e-5] + bool use_global_stats = 5; //optinal,by default true,testing mode + float moving_average_fraction = 6; //optinal,[default = .999]; + + WeightDef estimated_mean = 7; + WeightDef estimated_variance = 8; + + WeightDef scale = 9; + WeightDef bias = 10; +}; + +message ScaleOpParams { + WeightDef scale = 1; + WeightDef bias = 2; +}; + +message ReshapeOpParams { + float alpha = 1; + float beta = 2; + ShapeDef shape = 3; + int32 axis = 4; + int32 num_axes = 5; + int32 format = 6; +}; + +message SoftmaxOpParams { + int32 algo = 1; + int32 mode = 2; + float alpha = 3; + float beta = 4; +}; + +message FullConnectionOpParams { + WeightDef filter = 1; + WeightDef bias = 2; + uint32 num_output = 3; + bool relu_flag = 12; +}; + +message FlattenOpParams { + float alpha = 1; + float beta = 2; + int32 start_axis = 3; + int32 end_axis = 4; +} + +message AddLimitedOpParams { + float alpha = 1; + float beta = 2; + int32 axis = 3; + bool broadcast = 4; + + repeated WeightDef weight = 10; +}; + +message MulLimitedOpParams { + float alpha = 1; + float beta = 2; + int32 axis = 3; + bool broadcast = 4; + + repeated WeightDef weight = 10; +}; + +message AddOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message MulOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message SubOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message BiasAddOpParams { + float alpha = 1; + float beta = 2; + + WeightDef bias = 10; +}; + +message MatMulOpParams { + float alpha = 1; + float beta = 2; + bool transposeX = 3; + bool transposeW = 4; + + WeightDef filter = 10; + WeightDef bias = 12; +}; + +message RsqrtOpParams { + float alpha = 1; + float beta = 2; +}; + + +message WeightDef { + int32 format = 1; + int32 data_type = 2; + ShapeDef shape = 3; + bytes data = 4; + int64 data_offset = 5; + uint32 cmps_size = 6; + bytes cmps_tab = 7; + int64 cmps_tab_offset = 10; + CompressInfo cmps_info = 8; + AllOffsetQuantizeInfo alloffset_quantize_info = 11; +} + +message ShapeDef { + repeated int64 dim = 1; +} + +enum DeviceType { + NPU = 0; // In default, we will use NPU. + CPU = 1; // CPU +} + +message AllOffsetQuantizeInfo { + float scale = 1; + int32 offset = 2; +} + +message TensorDescriptor { + int32 format = 1; + int32 data_type = 2; + repeated int64 dim = 3; + uint32 size = 4; + bool reuse_input = 5; + bool output_tensor = 7; + DeviceType device_type = 8; + bool input_tensor = 9; + uint32 real_dim_cnt = 10; + uint32 reuse_input_index = 11; + AllOffsetQuantizeInfo alloffset_quantize_info = 12; +} + +message CompressInfo { + int32 blockRow = 1; // block row + int32 blockCol = 2; // block col + int32 fractalK = 3; // fractal K + int32 fractalN = 4; // fractal N + int32 lastFractalK = 5; // K of last fractal + int32 lastFractalN = 6; // N of last fractal + int32 cubeSize = 7; // cube's length + int32 loadDir = 8; // data load directtiono 0:col load 1:row load +} + +message AttrDef { + message ListValue { + repeated string s = 2; // "list(string)" + repeated int64 i = 3 [packed = true]; // "list(int)" + repeated float f = 4 [packed = true]; // "list(float)" + repeated bool b = 5 [packed = true]; // "list(bool)" + repeated uint32 u = 6 [packed = true]; // "list(uint)" + repeated bytes bt = 7; + } + + oneof value { + string s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + uint32 u = 6; // "uint32" + bytes bt = 7; + ListValue list = 1; // any "list(...)" + NamedAttrs func = 10; + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NamedAttrs { + string name = 1; + map attr = 2; +} + diff --git a/ge/common/proto/op_mapping_info.proto b/ge/common/proto/op_mapping_info.proto new file mode 100644 index 00000000..e23b7ebe --- /dev/null +++ b/ge/common/proto/op_mapping_info.proto @@ -0,0 +1,73 @@ +syntax = "proto3"; +package aicpu.dump; + +message Shape { + repeated uint64 dim = 1; +} + +message Output { + int32 data_type = 1; + int32 format = 2; + Shape shape = 3; + uint64 address = 4; + string original_name = 5; + int32 original_output_index = 6; + int32 original_output_data_type = 7; + int32 original_output_format = 8; + uint64 size = 9; +} + +message Input { + int32 data_type =1; + int32 format = 2; + Shape shape = 3; + uint64 address = 4; + uint64 size = 5; +} + +enum BufferType { + L1 = 0; +} + +message OpBuffer { + BufferType buffer_type = 1; + uint64 address = 2; + uint64 size = 3; +} + +message Op { + string op_name = 1; + string op_type = 2; +} + +message Task { + uint32 task_id = 1; + uint32 stream_id = 2; + Op op = 3; + repeated Output output = 4; + bool end_graph = 5; + repeated Input input = 6; + repeated OpBuffer buffer = 7; +} + +message OpMappingInfo { + string dump_path = 1; + oneof model_name_param { + string model_name = 2; + } + oneof model_id_param { + uint32 model_id = 3; + } + oneof step_id { + uint64 step_id_addr = 4; + } + oneof iterations_per_loop { + uint64 iterations_per_loop_addr = 5; + } + oneof loop_cond { + uint64 loop_cond_addr = 6; + } + uint32 flag = 7; // 0x01 load, 0x00 unload + repeated Task task = 8; + string dump_step = 9; +} \ No newline at end of file diff --git a/ge/common/proto/task.proto b/ge/common/proto/task.proto new file mode 100644 index 00000000..d0c09840 --- /dev/null +++ b/ge/common/proto/task.proto @@ -0,0 +1,165 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +package domi; + +message ModelTaskDef { + string version = 1; + + map attr = 9; // Extended field + repeated TaskDef task = 10; + + uint64 memory_size = 11; + uint32 stream_num = 12; + uint32 event_num = 13; + uint64 weight_size = 14; + + repeated bytes op = 15; // input/output opdef in bytes + + uint64 base_addr = 16; // base addr + uint64 weight_addr = 17; // weight addr + uint32 batch_num = 18; +} + + +message TaskDef { + uint32 id = 1; + uint32 type = 2; + + uint32 stream_id = 10; + uint32 event_id = 11; + + KernelDef kernel = 20; + KernelExDef kernel_ex = 21; + KernelHcclDef kernel_hccl = 25; + EventExDef event_ex = 26; + LogTimeStampDef log_timestamp = 28; + + uint32 label_id = 30; + + MemcpyAsyncDef memcpy_async = 31; + StreamSwitchDef stream_switch = 32; + StreamActiveDef stream_active = 33; + bytes private_def = 34; + uint64 ops_kernel_store_ptr = 35; // adjustments to other fields in the future + StreamSwitchNDef stream_switch_n = 36; + + LabelSetDef label_set = 37; + LabelGotoExDef label_goto_ex = 38; + LabelSwitchByIndexDef label_switch_by_index = 39; +} + +message KernelDef { + KernelContext context = 1; + + string stub_func = 10; + uint32 block_dim = 11; + uint32 args_size = 12; + bytes args = 13; + bytes sm_desc = 14; + bytes flowtable = 15; + string so_name = 16; + string kernel_name = 17; + bytes kernel_ext_info = 18; + uint32 kernel_ext_info_size = 19; +} + +message KernelContext { + uint32 kernel_type = 1; + uint32 op_id = 2; // OP type in CCE + uint32 kernel_func_id = 3; + uint32 op_index = 4; // TE/Custom operator + bool is_flowtable = 5; // Identify whether args is a flowtable structure + bytes args_offset = 6; // args offset information + uint32 args_count = 7; // args count + repeated uint32 origin_op_index = 8; +} + + +message KernelExDef { + uint32 flags = 1; + + uint32 op_index = 4; + uint32 args_size = 12; + bytes args = 13; + bytes task_info = 14; // serialized nodeDef, funcDef, inputoutput + uint32 task_info_size = 15; + bytes kernel_ext_info = 16; + uint32 kernel_ext_info_size = 17; +} + + +message KernelHcclDef { + uint32 op_index = 8; + string hccl_type = 9; +} + + +message EventExDef { + uint32 op_index = 1; + uint32 event_type = 2; +} + +message LogTimeStampDef { + uint64 logid = 1; + bool notify = 2; + uint32 flat = 3; +} + +message MemcpyAsyncDef { + uint64 dst = 1; + uint64 dst_max = 2; + uint64 src = 3; + uint64 count = 4; + uint32 kind = 5; + uint32 op_index = 6; +} + +message StreamSwitchDef { + uint32 op_index = 1; + uint32 true_stream_id = 2; + int64 value = 3; + uint64 value_ptr = 4; + uint32 data_type = 5; +} + +message StreamActiveDef { + uint32 op_index = 1; + uint32 active_stream_id = 2; +} + +message StreamSwitchNDef { + uint32 op_index = 1; + uint32 size = 2; + repeated int64 target_value = 3; + repeated uint32 true_stream_id = 4; + uint32 element_size = 5; + uint32 data_type = 6; +} + +message LabelSetDef { + uint32 op_index = 1; + uint32 label_id = 2; + uint32 model_id = 3; +} + +message LabelGotoExDef { + uint32 op_index = 1; + uint32 label_id = 2; + uint32 model_id = 3; +} + +message LabelSwitchByIndexDef { + uint32 op_index = 1; + uint32 label_max = 2; +} diff --git a/ge/common/proto/tensorflow/attr_value.proto b/ge/common/proto/tensorflow/attr_value.proto new file mode 100644 index 00000000..1cc67d62 --- /dev/null +++ b/ge/common/proto/tensorflow/attr_value.proto @@ -0,0 +1,62 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "AttrValueProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "tensor.proto"; +import "tensor_shape.proto"; +import "types.proto"; + +// Protocol buffer representing the value for an attr used to configure an Op. +// Comment indicates the corresponding attr type. Only the field matching the +// attr type may be filled. +message AttrValue { + // LINT.IfChange + message ListValue { + repeated bytes s = 2; // "list(string)" + repeated int64 i = 3 [packed = true]; // "list(int)" + repeated float f = 4 [packed = true]; // "list(float)" + repeated bool b = 5 [packed = true]; // "list(bool)" + repeated DataType type = 6 [packed = true]; // "list(type)" + repeated TensorShapeProto shape = 7; // "list(shape)" + repeated TensorProto tensor = 8; // "list(tensor)" + repeated NameAttrList func = 9; // "list(attr)" + } + // LINT.ThenChange(https://www.tensorflow.org/code/tensorflow/c/c_api.cc) + + oneof value { + bytes s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + DataType type = 6; // "type" + TensorShapeProto shape = 7; // "shape" + TensorProto tensor = 8; // "tensor" + ListValue list = 1; // any "list(...)" + + // "func" represents a function. func.name is a function's name or + // a primitive op's name. func.attr.first is the name of an attr + // defined for that function. func.attr.second is the value for + // that attr in the instantiation. + NameAttrList func = 10; + + // This is a placeholder only used in nodes defined inside a + // function. It indicates the attr value will be supplied when + // the function is instantiated. For example, let us suppose a + // node "N" in function "FN". "N" has an attr "A" with value + // placeholder = "foo". When FN is instantiated with attr "foo" + // set to "bar", the instantiated node N's attr A will have been + // given the value "bar". + string placeholder = 9; + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NameAttrList { + string name = 1; + map attr = 2; +} diff --git a/ge/common/proto/tensorflow/function.proto b/ge/common/proto/tensorflow/function.proto new file mode 100644 index 00000000..075897c6 --- /dev/null +++ b/ge/common/proto/tensorflow/function.proto @@ -0,0 +1,100 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "FunctionProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "attr_value.proto"; +import "node_def.proto"; +import "op_def.proto"; + +// A library is a set of named functions. +message FunctionDefLibrary { + repeated FunctionDef function = 1; + repeated GradientDef gradient = 2; +} + +// A function can be instantiated when the runtime can bind every attr +// with a value. When a GraphDef has a call to a function, it must +// have binding for every attr defined in the signature. +// * device spec, etc. +message FunctionDef { + // The definition of the function's name, arguments, return values, + // attrs etc. + OpDef signature = 1; + + // Attributes specific to this function definition. + map attr = 5; + + // NOTE: field id 2 deleted on Jan 11, 2017, GraphDef version 21. + reserved 2; + + // In both of the following fields, there is the need to specify an + // output that is used as either the input to another node (in + // `node_def`) or as a return value of the function (in `ret`). + // Unlike the NodeDefs in GraphDef, we need to be able to specify a + // list in some cases (instead of just single outputs). Also, we + // need to be able to deal with lists of unknown length (so the + // output index may not be known at function definition time). So + // we use the following format instead: + // * "fun_in" where "fun_in" is the name of a function input arg in + // the `signature` field above. This represents that input, whether + // it is a single tensor or a list. + // * "fun_in:0" gives the first element of a function input arg (a + // non-list input is considered a list of length 1 for these + // purposes). + // * "node:out" where "node" is the name of a node in `node_def` and + // "out" is the name one of its op's output arguments (the name + // comes from the OpDef of the node's op). This represents that + // node's output, whether it is a single tensor or a list. + // Note: We enforce that an op's output arguments are never + // renamed in the backwards-compatibility test. + // * "node:out:0" gives the first element of a node output arg (a + // non-list output is considered a list of length 1 for these + // purposes). + // + // NOT CURRENTLY SUPPORTED (but may be in the future): + // * "node:out:-1" gives last element in a node output list + // * "node:out:1:" gives a list with all but the first element in a + // node output list + // * "node:out::-1" gives a list with all but the last element in a + // node output list + + // The body of the function. Unlike the NodeDefs in a GraphDef, attrs + // may have values of type `placeholder` and the `input` field uses + // the "output" format above. + + // By convention, "op" in node_def is resolved by consulting with a + // user-defined library first. If not resolved, "func" is assumed to + // be a builtin op. + repeated NodeDef node_def = 3; + + // A mapping from the output arg names from `signature` to the + // outputs from `node_def` that should be returned by the function. + map ret = 4; +} + +// GradientDef defines the gradient function of a function defined in +// a function library. +// +// A gradient function g (specified by gradient_func) for a function f +// (specified by function_name) must follow the following: +// +// The function 'f' must be a numerical function which takes N inputs +// and produces M outputs. Its gradient function 'g', which is a +// function taking N + M inputs and produces N outputs. +// +// I.e. if we have +// (y1, y2, ..., y_M) = f(x1, x2, ..., x_N), +// then, g is +// (dL/dx1, dL/dx2, ..., dL/dx_N) = g(x1, x2, ..., x_N, +// dL/dy1, dL/dy2, ..., dL/dy_M), +// where L is a scalar-value function of (x1, x2, ..., xN) (e.g., the +// loss function). dL/dx_i is the partial derivative of L with respect +// to x_i. +message GradientDef { + string function_name = 1; // The function name. + string gradient_func = 2; // The gradient function's name. +} diff --git a/ge/common/proto/tensorflow/graph.proto b/ge/common/proto/tensorflow/graph.proto new file mode 100644 index 00000000..d639a7d6 --- /dev/null +++ b/ge/common/proto/tensorflow/graph.proto @@ -0,0 +1,56 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "GraphProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "node_def.proto"; +import "function.proto"; +import "versions.proto"; + +// Represents the graph of operations +message GraphDef { + repeated NodeDef node = 1; + + // Compatibility versions of the graph. See core/public/version.h for version + // history. The GraphDef version is distinct from the TensorFlow version, and + // each release of TensorFlow will support a range of GraphDef versions. + VersionDef versions = 4; + + // Deprecated single version field; use versions above instead. Since all + // GraphDef changes before "versions" was introduced were forward + // compatible, this field is entirely ignored. + int32 version = 3 [deprecated = true]; + + // EXPERIMENTAL. DO NOT USE OR DEPEND ON THIS YET. + // + // "library" provides user-defined functions. + // + // Naming: + // * library.function.name are in a flat namespace. + // NOTE: We may need to change it to be hierarchical to support + // different orgs. E.g., + // { "/google/nn", { ... }}, + // { "/google/vision", { ... }} + // { "/org_foo/module_bar", { ... }} + // map named_lib; + // * If node[i].op is the name of one function in "library", + // node[i] is deemed as a function call. Otherwise, node[i].op + // must be a primitive operation supported by the runtime. + // + // + // Function call semantics: + // + // * The callee may start execution as soon as some of its inputs + // are ready. The caller may want to use Tuple() mechanism to + // ensure all inputs are ready in the same time. + // + // * The consumer of return values may start executing as soon as + // the return values the consumer depends on are ready. The + // consumer may want to use Tuple() mechanism to ensure the + // consumer does not start until all return values of the callee + // function are ready. + FunctionDefLibrary library = 2; +}; diff --git a/ge/common/proto/tensorflow/graph_library.proto b/ge/common/proto/tensorflow/graph_library.proto new file mode 100644 index 00000000..e393d38d --- /dev/null +++ b/ge/common/proto/tensorflow/graph_library.proto @@ -0,0 +1,14 @@ +syntax = "proto3"; + +package domi.tensorflow; + +import "graph.proto"; + +message GeGraphDef { + string name = 1; + GraphDef graph = 2; +} + +message GraphDefLibrary { + repeated GeGraphDef graph_def = 1; +}; \ No newline at end of file diff --git a/ge/common/proto/tensorflow/node_def.proto b/ge/common/proto/tensorflow/node_def.proto new file mode 100644 index 00000000..b9bc97ee --- /dev/null +++ b/ge/common/proto/tensorflow/node_def.proto @@ -0,0 +1,63 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "NodeProto"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "attr_value.proto"; + +message NodeDef { + // The name given to this operator. Used for naming inputs, + // logging, visualization, etc. Unique within a single GraphDef. + // Must match the regexp "[A-Za-z0-9.][A-Za-z0-9_./]*". + string name = 1; + + // The operation name. There may be custom parameters in attrs. + // Op names starting with an underscore are reserved for internal use. + string op = 2; + + // Each input is "node:src_output" with "node" being a string name and + // "src_output" indicating which output tensor to use from "node". If + // "src_output" is 0 the ":0" suffix can be omitted. Regular inputs + // may optionally be followed by control inputs that have the format + // "^node". + repeated string input = 3; + + // A (possibly partial) specification for the device on which this + // node should be placed. + // The expected syntax for this string is as follows: + // + // DEVICE_SPEC ::= PARTIAL_SPEC + // + // PARTIAL_SPEC ::= ("/" CONSTRAINT) * + // CONSTRAINT ::= ("job:" JOB_NAME) + // | ("replica:" [1-9][0-9]*) + // | ("task:" [1-9][0-9]*) + // | ("device:" [A-Za-z]* ":" ([1-9][0-9]* | "*") ) + // + // Valid values for this string include: + // * "/job:worker/replica:0/task:1/device:GPU:3" (full specification) + // * "/job:worker/device:GPU:3" (partial specification) + // * "" (no specification) + // + // If the constraints do not resolve to a single device (or if this + // field is empty or not present), the runtime will attempt to + // choose a device automatically. + string device = 4; + + // Operation-specific graph-construction-time configuration. + // Note that this should include all attrs defined in the + // corresponding OpDef, including those with a value matching + // the default -- this allows the default to change and makes + // NodeDefs easier to interpret on their own. However, if + // an attr with a default is not specified in this list, the + // default will be used. + // The "names" (keys) must match the regexp "[a-z][a-z0-9_]+" (and + // one of the names from the corresponding OpDef's attr field). + // The values must have a type matching the corresponding OpDef + // attr's type field. + // Add some examples here showing best practices. + map attr = 5; +}; diff --git a/ge/common/proto/tensorflow/op_def.proto b/ge/common/proto/tensorflow/op_def.proto new file mode 100644 index 00000000..3485d045 --- /dev/null +++ b/ge/common/proto/tensorflow/op_def.proto @@ -0,0 +1,164 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "OpDefProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "attr_value.proto"; +import "types.proto"; + +// Defines an operation. A NodeDef in a GraphDef specifies an Op by +// using the "op" field which should match the name of a OpDef. +// LINT.IfChange +message OpDef { + // Op names starting with an underscore are reserved for internal use. + // Names should be CamelCase and match the regexp "[A-Z][a-zA-Z0-9_]*". + string name = 1; + + // For describing inputs and outputs. + message ArgDef { + // Name for the input/output. Should match the regexp "[a-z][a-z0-9_]*". + string name = 1; + + // Human readable description. + string description = 2; + + // Describes the type of one or more tensors that are accepted/produced + // by this input/output arg. The only legal combinations are: + // * For a single tensor: either the "type" field is set or the + // "type_attr" field is set to the name of an attr with type "type". + // * For a sequence of tensors with the same type: the "number_attr" + // field will be set to the name of an attr with type "int", and + // either the "type" or "type_attr" field will be set as for + // single tensors. + // * For a sequence of tensors, the "type_list_attr" field will be set + // to the name of an attr with type "list(type)". + DataType type = 3; + string type_attr = 4; // if specified, attr must have type "type" + string number_attr = 5; // if specified, attr must have type "int" + // If specified, attr must have type "list(type)", and none of + // type, type_attr, and number_attr may be specified. + string type_list_attr = 6; + + // For inputs: if true, the inputs are required to be refs. + // By default, inputs can be either refs or non-refs. + // For outputs: if true, outputs are refs, otherwise they are not. + bool is_ref = 16; + }; + + // Description of the input(s). + repeated ArgDef input_arg = 2; + + // Description of the output(s). + repeated ArgDef output_arg = 3; + + // Description of the graph-construction-time configuration of this + // Op. That is to say, this describes the attr fields that will + // be specified in the NodeDef. + message AttrDef { + // A descriptive name for the argument. May be used, e.g. by the + // Python client, as a keyword argument name, and so should match + // the regexp "[a-z][a-z0-9_]+". + string name = 1; + + // One of the type names from attr_value.proto ("string", "list(string)", + // "int", etc.). + string type = 2; + + // A reasonable default for this attribute if the user does not supply + // a value. If not specified, the user must supply a value. + AttrValue default_value = 3; + + // Human-readable description. + string description = 4; + + + // --- Constraints --- + // These constraints are only in effect if specified. Default is no + // constraints. + + // For type == "int", this is a minimum value. For "list(___)" + // types, this is the minimum length. + bool has_minimum = 5; + int64 minimum = 6; + + // The set of allowed values. Has type that is the "list" version + // of the "type" field above (uses the "list" field of AttrValue). + // If type == "type" or "list(type)" above, then the "type" field + // of "allowed_values.list" has the set of allowed DataTypes. + // If type == "string" or "list(string)", then the "s" field of + // "allowed_values.list" has the set of allowed strings. + AttrValue allowed_values = 7; + } + repeated AttrDef attr = 4; + + // Optional deprecation based on GraphDef versions. + OpDeprecation deprecation = 8; + + // One-line human-readable description of what the Op does. + string summary = 5; + + // Additional, longer human-readable description of what the Op does. + string description = 6; + + // ------------------------------------------------------------------------- + // Which optimizations this operation can participate in. + + // True if the operation is commutative ("op(a,b) == op(b,a)" for all inputs) + bool is_commutative = 18; + + // If is_aggregate is true, then this operation accepts N >= 2 + // inputs and produces 1 output all of the same type. Should be + // associative and commutative, and produce output with the same + // shape as the input. The optimizer may replace an aggregate op + // taking input from multiple devices with a tree of aggregate ops + // that aggregate locally within each device (and possibly within + // groups of nearby devices) before communicating. + bool is_aggregate = 16; // for things like add + + // Other optimizations go here, like + // can_alias_input, rewrite_when_output_unused, partitioning_strategy, etc. + + // ------------------------------------------------------------------------- + // Optimization constraints. + + // Ops are marked as stateful if their behavior depends on some state beyond + // their input tensors (e.g. variable reading op) or if they have + // a side-effect (e.g. printing or asserting ops). Equivalently, stateless ops + // must always produce the same output for the same input and have + // no side-effects. + // + // By default Ops may be moved between devices. Stateful ops should + // either not be moved, or should only be moved if that state can also + // be moved (e.g. via some sort of save / restore). + // Stateful ops are guaranteed to never be optimized away by Common + // Subexpression Elimination (CSE). + bool is_stateful = 17; // for things like variables, queue + + // ------------------------------------------------------------------------- + // Non-standard options. + + // By default, all inputs to an Op must be initialized Tensors. Ops + // that may initialize tensors for the first time should set this + // field to true, to allow the Op to take an uninitialized Tensor as + // input. + bool allows_uninitialized_input = 19; // for Assign, etc. +}; +// LINT.ThenChange( +// https://www.tensorflow.org/code/tensorflow/core/framework/op_def_util.cc) + +// Information about version-dependent deprecation of an op +message OpDeprecation { + // First GraphDef version at which the op is disallowed. + int32 version = 1; + + // Explanation of why it was deprecated and what to use instead. + string explanation = 2; +}; + +// A collection of OpDefs +message OpList { + repeated OpDef op = 1; +}; diff --git a/ge/common/proto/tensorflow/resource_handle.proto b/ge/common/proto/tensorflow/resource_handle.proto new file mode 100644 index 00000000..a3452351 --- /dev/null +++ b/ge/common/proto/tensorflow/resource_handle.proto @@ -0,0 +1,29 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "ResourceHandle"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +// Protocol buffer representing a handle to a tensorflow resource. Handles are +// not valid across executions, but can be serialized back and forth from within +// a single run. +message ResourceHandleProto { + // Unique name for the device containing the resource. + string device = 1; + + // Container in which this resource is placed. + string container = 2; + + // Unique name of this resource. + string name = 3; + + // Hash code for the type of the resource. Is only valid in the same device + // and in the same execution. + uint64 hash_code = 4; + + // For debug-only, the name of the type pointed to by this handle, if + // available. + string maybe_type_name = 5; +}; diff --git a/ge/common/proto/tensorflow/tensor.proto b/ge/common/proto/tensorflow/tensor.proto new file mode 100644 index 00000000..d0a4d024 --- /dev/null +++ b/ge/common/proto/tensorflow/tensor.proto @@ -0,0 +1,94 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "TensorProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "resource_handle.proto"; +import "tensor_shape.proto"; +import "types.proto"; + +// Protocol buffer representing a tensor. +message TensorProto { + DataType dtype = 1; + + // Shape of the tensor. + TensorShapeProto tensor_shape = 2; + + // Only one of the representations below is set, one of "tensor_contents" and + // the "xxx_val" attributes. We are not using oneof because as oneofs cannot + // contain repeated fields it would require another extra set of messages. + + // Version number. + // + // In version 0, if the "repeated xxx" representations contain only one + // element, that element is repeated to fill the shape. This makes it easy + // to represent a constant Tensor with a single value. + int32 version_number = 3; + + // Serialized raw tensor content from either Tensor::AsProtoTensorContent or + // memcpy in tensorflow::grpc::EncodeTensorToByteBuffer. This representation + // can be used for all tensor types. The purpose of this representation is to + // reduce serialization overhead during RPC call by avoiding serialization of + // many repeated small items. + bytes tensor_content = 4; + + // Type specific representations that make it easy to create tensor protos in + // all languages. Only the representation corresponding to "dtype" can + // be set. The values hold the flattened representation of the tensor in + // row major order. + + // DT_HALF, DT_BFLOAT16. Note that since protobuf has no int16 type, we'll + // have some pointless zero padding for each value here. + repeated int32 half_val = 13 [packed = true]; + + // DT_FLOAT. + repeated float float_val = 5 [packed = true]; + + // DT_DOUBLE. + repeated double double_val = 6 [packed = true]; + + // DT_INT32, DT_INT16, DT_INT8, DT_UINT8. + repeated int32 int_val = 7 [packed = true]; + + // DT_STRING + repeated bytes string_val = 8; + + // DT_COMPLEX64. scomplex_val(2*i) and scomplex_val(2*i+1) are real + // and imaginary parts of i-th single precision complex. + repeated float scomplex_val = 9 [packed = true]; + + // DT_INT64 + repeated int64 int64_val = 10 [packed = true]; + + // DT_BOOL + repeated bool bool_val = 11 [packed = true]; + + // DT_COMPLEX128. dcomplex_val(2*i) and dcomplex_val(2*i+1) are real + // and imaginary parts of i-th double precision complex. + repeated double dcomplex_val = 12 [packed = true]; + + // DT_RESOURCE + repeated ResourceHandleProto resource_handle_val = 14; + + // DT_VARIANT + repeated VariantTensorDataProto variant_val = 15; + + // DT_UINT32 + repeated uint32 uint32_val = 16 [packed = true]; + + // DT_UINT64 + repeated uint64 uint64_val = 17 [packed = true]; +}; + +// Protocol buffer representing the serialization format of DT_VARIANT tensors. +message VariantTensorDataProto { + // Name of the type of objects being serialized. + string type_name = 1; + // Portions of the object that are not Tensors. + bytes metadata = 2; + // Tensors contained within objects being serialized. + repeated TensorProto tensors = 3; +} diff --git a/ge/common/proto/tensorflow/tensor_shape.proto b/ge/common/proto/tensorflow/tensor_shape.proto new file mode 100644 index 00000000..4225a2e3 --- /dev/null +++ b/ge/common/proto/tensorflow/tensor_shape.proto @@ -0,0 +1,45 @@ +// Protocol buffer representing the shape of tensors. + +syntax = "proto3"; +option cc_enable_arenas = true; +option java_outer_classname = "TensorShapeProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +package domi.tensorflow; + +// Dimensions of a tensor. +message TensorShapeProto { + // One dimension of the tensor. + message Dim { + // Size of the tensor in that dimension. + // This value must be >= -1, but values of -1 are reserved for "unknown" + // shapes (values of -1 mean "unknown" dimension). Certain wrappers + // that work with TensorShapeProto may fail at runtime when deserializing + // a TensorShapeProto containing a dim value of -1. + int64 size = 1; + + // Optional name of the tensor dimension. + string name = 2; + }; + + // Dimensions of the tensor, such as {"input", 30}, {"output", 40} + // for a 30 x 40 2D tensor. If an entry has size -1, this + // corresponds to a dimension of unknown size. The names are + // optional. + // + // The order of entries in "dim" matters: It indicates the layout of the + // values in the tensor in-memory representation. + // + // The first entry in "dim" is the outermost dimension used to layout the + // values, the last entry is the innermost dimension. This matches the + // in-memory layout of RowMajor Eigen tensors. + // + // If "dim.size()" > 0, "unknown_rank" must be false. + repeated Dim dim = 2; + + // If true, the number of dimensions in the shape is unknown. + // + // If true, "dim.size()" must be 0. + bool unknown_rank = 3; +}; diff --git a/ge/common/proto/tensorflow/types.proto b/ge/common/proto/tensorflow/types.proto new file mode 100644 index 00000000..ba7a72b3 --- /dev/null +++ b/ge/common/proto/tensorflow/types.proto @@ -0,0 +1,74 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "TypesProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +// LINT.IfChange +enum DataType { + // Not a legal value for DataType. Used to indicate a DataType field + // has not been set. + DT_INVALID = 0; + + // Data types that all computation devices are expected to be + // capable to support. + DT_FLOAT = 1; + DT_DOUBLE = 2; + DT_INT32 = 3; + DT_UINT8 = 4; + DT_INT16 = 5; + DT_INT8 = 6; + DT_STRING = 7; + DT_COMPLEX64 = 8; // Single-precision complex + DT_INT64 = 9; + DT_BOOL = 10; + DT_QINT8 = 11; // Quantized int8 + DT_QUINT8 = 12; // Quantized uint8 + DT_QINT32 = 13; // Quantized int32 + DT_BFLOAT16 = 14; // Float32 truncated to 16 bits. Only for cast ops. + DT_QINT16 = 15; // Quantized int16 + DT_QUINT16 = 16; // Quantized uint16 + DT_UINT16 = 17; + DT_COMPLEX128 = 18; // Double-precision complex + DT_HALF = 19; + DT_RESOURCE = 20; + DT_VARIANT = 21; // Arbitrary C++ data types + DT_UINT32 = 22; + DT_UINT64 = 23; + + // Do not use! These are only for parameters. Every enum above + // should have a corresponding value below (verified by types_test). + DT_FLOAT_REF = 101; + DT_DOUBLE_REF = 102; + DT_INT32_REF = 103; + DT_UINT8_REF = 104; + DT_INT16_REF = 105; + DT_INT8_REF = 106; + DT_STRING_REF = 107; + DT_COMPLEX64_REF = 108; + DT_INT64_REF = 109; + DT_BOOL_REF = 110; + DT_QINT8_REF = 111; + DT_QUINT8_REF = 112; + DT_QINT32_REF = 113; + DT_BFLOAT16_REF = 114; + DT_QINT16_REF = 115; + DT_QUINT16_REF = 116; + DT_UINT16_REF = 117; + DT_COMPLEX128_REF = 118; + DT_HALF_REF = 119; + DT_RESOURCE_REF = 120; + DT_VARIANT_REF = 121; + DT_UINT32_REF = 122; + DT_UINT64_REF = 123; +} +// LINT.ThenChange( +// https://www.tensorflow.org/code/tensorflow/c/c_api.h, +// https://www.tensorflow.org/code/tensorflow/go/tensor.go, +// https://www.tensorflow.org/code/tensorflow/core/framework/tensor.cc, +// https://www.tensorflow.org/code/tensorflow/core/framework/types.h, +// https://www.tensorflow.org/code/tensorflow/core/framework/types.cc, +// https://www.tensorflow.org/code/tensorflow/python/framework/dtypes.py, +// https://www.tensorflow.org/code/tensorflow/python/framework/function.py) diff --git a/ge/common/proto/tensorflow/versions.proto b/ge/common/proto/tensorflow/versions.proto new file mode 100644 index 00000000..48061218 --- /dev/null +++ b/ge/common/proto/tensorflow/versions.proto @@ -0,0 +1,31 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "VersionsProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +// Version information for a piece of serialized data +// +// There are different types of versions for each type of data +// (GraphDef, etc.), but they all have the same common shape +// described here. +// +// Each consumer has "consumer" and "min_producer" versions (specified +// elsewhere). A consumer is allowed to consume this data if +// +// producer >= min_producer +// consumer >= min_consumer +// consumer not in bad_consumers +// +message VersionDef { + // The version of the code that produced this data. + int32 producer = 1; + + // Any consumer below this version is not allowed to consume this data. + int32 min_consumer = 2; + + // Specific consumer versions which are disallowed (e.g. due to bugs). + repeated int32 bad_consumers = 3; +}; diff --git a/ge/executor/proto/dump_task.proto b/ge/executor/proto/dump_task.proto new file mode 100644 index 00000000..b1e346cd --- /dev/null +++ b/ge/executor/proto/dump_task.proto @@ -0,0 +1,111 @@ +syntax = "proto3"; +package toolkit.dumpdata; + +enum OutputDataType { + DT_UNDEFINED = 0; + DT_FLOAT = 1; + DT_FLOAT16 = 2; + DT_INT8 = 3; + DT_UINT8 = 4; + DT_INT16 = 5; + DT_UINT16 = 6; + DT_INT32 = 7; + DT_INT64 = 8; + DT_UINT32 = 9; + DT_UINT64 = 10; + DT_BOOL = 11; + DT_DOUBLE = 12; + DT_STRING = 13; + DT_DUAL_SUB_INT8 = 14; + DT_DUAL_SUB_UINT8 = 15; + DT_COMPLEX64 = 16; + DT_COMPLEX128 = 17; + DT_QINT8 = 18; + DT_QINT16 = 19; + DT_QINT32 = 20; + DT_QUINT8 = 21; + DT_QUINT16 = 22; + DT_RESOURCE = 23; + DT_STRING_REF = 24; + DT_DUAL = 25; +} + +enum OutputFormat { + FORMAT_NCHW = 0; + FORMAT_NHWC = 1; + FORMAT_ND = 2; + FORMAT_NC1HWC0 = 3; + FORMAT_FRACTAL_Z = 4; + FORMAT_NC1C0HWPAD = 5; + FORMAT_NHWC1C0 = 6; + FORMAT_FSR_NCHW = 7; + FORMAT_FRACTAL_DECONV = 8; + FORMAT_C1HWNC0 = 9; + FORMAT_FRACTAL_DECONV_TRANSPOSE = 10; + FORMAT_FRACTAL_DECONV_SP_STRIDE_TRANS = 11; + FORMAT_NC1HWC0_C04 = 12; + FORMAT_FRACTAL_Z_C04 = 13; + FORMAT_CHWN = 14; + FORMAT_FRACTAL_DECONV_SP_STRIDE8_TRANS = 15; + FORMAT_HWCN = 16; + FORMAT_NC1KHKWHWC0 = 17; + FORMAT_BN_WEIGHT = 18; + FORMAT_FILTER_HWCK = 19; + FORMAT_HASHTABLE_LOOKUP_LOOKUPS=20; + FORMAT_HASHTABLE_LOOKUP_KEYS = 21; + FORMAT_HASHTABLE_LOOKUP_VALUE = 22; + FORMAT_HASHTABLE_LOOKUP_OUTPUT = 23; + FORMAT_HASHTABLE_LOOKUP_HITS=24; + FORMAT_C1HWNCoC0 = 25; + FORMAT_MD = 26; + FORMAT_NDHWC = 27; + FORMAT_FRACTAL_ZZ = 28; + FORMAT_FRACTAL_NZ = 29; + FORMAT_RESERVED = 30; +} + +message OriginalOp { + string name = 1; + uint32 output_index = 2; + OutputDataType data_type = 3; + OutputFormat format = 4; +} + +message Shape { + repeated uint64 dim = 1; +} + +message OpOutput { + OutputDataType data_type = 1; + OutputFormat format = 2; + Shape shape = 3; + OriginalOp original_op = 4; // the original op corresponding to the output + bytes data = 5; + uint64 size = 6; +} + +message OpInput { + OutputDataType data_type = 1; + OutputFormat format = 2; + Shape shape = 3; + bytes data = 4; + uint64 size = 5; +} + +enum BufferType { + L1 = 0; +} + +message OpBuffer { + BufferType buffer_type = 1; + bytes data = 2; + uint64 size = 3; +} + +message DumpData{ + string version = 1; + uint64 dump_time = 2; + repeated OpOutput output = 3; + repeated OpInput input = 4; + repeated OpBuffer buffer = 5; +} diff --git a/ge/executor/proto/ge_ir.proto b/ge/executor/proto/ge_ir.proto new file mode 100644 index 00000000..e7bfe0cb --- /dev/null +++ b/ge/executor/proto/ge_ir.proto @@ -0,0 +1,190 @@ +syntax = "proto3"; + +package ge.proto; + +enum DataType +{ + DT_UNDEFINED = 0; // Used to indicate a DataType field has not been set. + DT_FLOAT = 1; // float type + DT_FLOAT16 = 2; // fp16 type + DT_INT8 = 3; // int8 type + DT_UINT8 = 4; // uint8 type + DT_INT16 = 5; // int16 type + DT_UINT16 = 6; // uint16 type + DT_INT32 = 7; // + DT_INT64 = 8; // int64 type + DT_UINT32 = 9; // unsigned int32 + DT_UINT64 = 10; // unsigned int64 + DT_BOOL = 11; // bool type + DT_DOUBLE = 12; // double type + DT_STRING = 13; // string type + DT_DUAL_SUB_INT8 = 14; /**< dual output int8 type */ + DT_DUAL_SUB_UINT8 = 15; /**< dual output uint8 type */ + DT_COMPLEX64 = 16; // complex64 type + DT_COMPLEX128 = 17; // complex128 type + DT_QINT8 = 18; // qint8 type + DT_QINT16 = 19; // qint16 type + DT_QINT32 = 20; // qint32 type + DT_QUINT8 = 21; // quint8 type + DT_QUINT16 = 22; // quint16 type + DT_RESOURCE = 23; // resource type + DT_STRING_REF = 24; // string_ref type + DT_DUAL = 25; /**< dual output type */ +} + +message AttrDef +{ + message ListValue + { + enum ListValueType{ + VT_LIST_NONE = 0; + VT_LIST_STRING = 1; + VT_LIST_INT = 2; + VT_LIST_FLOAT = 3; + VT_LIST_BOOL = 4; + VT_LIST_BYTES = 5; + VT_LIST_TENSOR_DESC = 6; + VT_LIST_TENSOR = 7; + VT_LIST_GRAPH = 8; + VT_LIST_NAMED_ATTRS = 9; + VT_LIST_DATA_TYPE = 10; + } + repeated bytes s = 2; // "list(string)" + repeated int64 i = 3; // "list(int)" + repeated float f = 4; // "list(float)" + repeated bool b = 5; // "list(bool)" + repeated bytes bt = 7; + repeated TensorDescriptor td = 8; + repeated TensorDef t = 9; + repeated GraphDef g = 10; + repeated NamedAttrs na = 11; + repeated int64 dt = 12; // list ge::DataType + + ListValueType val_type = 20; + } + + message ListListInt{ + message ListInt{ + repeated int64 list_i = 1; // list int + } + repeated ListInt list_list_i = 1; // list list int + } + + oneof value + { + bytes s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + bytes bt = 7; + ListValue list = 1; // any "list(...)" + NamedAttrs func = 10; // Used to support attr nesting + TensorDescriptor td = 11; // GeTensorDesc type + TensorDef t = 12; // GeTensor type + GraphDef g = 13; // Graph type + ListListInt list_list_int = 14; // List List Int type + int64 dt = 15; // ge::DataType + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NamedAttrs +{ + string name = 1; + map attr = 2; +} + +// Shape / dimension description, using row-major order +message ShapeDef +{ + repeated int64 dim = 1; // Size of each dimension +} + +// Multidimensional data description +message TensorDescriptor +{ + string name = 1; // Optional parameter, tensor name + + DataType dtype = 2; // tensor datatype + ShapeDef shape = 3; // Shape / dimension + string layout = 4; // Tensor format, eg: "NCHW", "NHWC", "CHW", "ND" + + bool has_out_attr = 9; + int64 size = 10; + int64 weight_size = 11; + bool reuse_input = 12; + bool output_tensor = 13; + string device_type = 14; + bool input_tensor =15; + int64 real_dim_cnt = 16; + int64 reuse_input_index = 17; + int64 data_offset = 18; + int64 cmps_size = 19; + string cmps_tab = 20; + int64 cmps_tab_offset = 21; + + map attr = 5; // Set of extra parameter fields +} + +// GeTensor definition +message TensorDef +{ + TensorDescriptor desc = 1; // Tensor description + bytes data = 2; // Tensor data +} + + +// Operator description +message OpDef +{ + string name = 1; // name + string type = 2; // type + + repeated string input = 5; // input original op name + outgoing index. op_name锛歩ndex + + map attr = 10; // Set of operator parameter fields + + bool has_out_attr = 20; + int64 id = 21; + int64 stream_id =22; + repeated string input_name = 23; + repeated string src_name = 24; + repeated int64 src_index = 25; + repeated string dst_name = 26; + repeated int64 dst_index = 27; + repeated int64 input_i = 28; + repeated int64 output_i = 29; + repeated int64 workspace = 30; + repeated int64 workspace_bytes = 31; + repeated bool is_input_const = 32; + repeated TensorDescriptor input_desc = 33; + repeated TensorDescriptor output_desc = 34; + repeated string subgraph_name = 35; +} + +// Graph definition +message GraphDef +{ + string name = 1; // name + + repeated string input = 4; // Graph input + repeated string output = 5; // Graph output + + repeated OpDef op = 6; // List of operators + + map attr = 11; // Extended field +} + +// model definition +message ModelDef +{ + string name = 1; // name + uint32 version = 2; // IR Proto verion + string custom_version = 3; // User model version number, passed in by user + + repeated GraphDef graph = 7; // Graph definition锛実raph[0] represents the main diagram in modeldef + + map attr = 11; // Extended field +} + diff --git a/ge/executor/proto/insert_op.proto b/ge/executor/proto/insert_op.proto new file mode 100644 index 00000000..c635ca14 --- /dev/null +++ b/ge/executor/proto/insert_op.proto @@ -0,0 +1,136 @@ +syntax = "proto3"; + +package domi; + +message InsertNewOps { + repeated AippOpParams aipp_op = 1; + repeated MultiShapeOpParams multi_shape_op = 2; +} + +message AippOpParams { + enum InputFormat { + UNDEFINED = 0; + YUV420SP_U8 = 1; + XRGB8888_U8 = 2; + RGB888_U8 = 3; + YUV400_U8 = 4; + NC1HWC0DI_FP16 = 5; + NC1HWC0DI_S8 = 6; + ARGB8888_U8 = 7; + YUYV_U8 = 8; + YUV422SP_U8 = 9; + AYUV444_U8 = 10; + RAW10 = 11; + RAW12 = 12; + RAW16 = 13; + RAW24 = 14; + RGB16 = 15; + RGB20 = 16; + RGB24 = 17; + RGB8_IR = 18; + RGB16_IR = 19; + RGB24_IR = 20; + } + + enum AippMode { + undefined = 0; + static = 1; + dynamic = 2; + } + + // AIPP模式，区分静态AIPP和动态AIPP + AippMode aipp_mode = 1; + + // related_input_rank参数为必填，类型为整型，配置范围>=0, <=输入Data算子的个数，默认值为0。 + // 标识对模型的第几个输入做AIPP处理，例如模型有两个输入，需要对第2个输入做AIPP，则配置related_input_rank为1。 + uint32 related_input_rank = 2; + + // input_edge_idx参数为可选，类型为整型，配置范围为>=0。 + // 配置该参数的作用，在于对Data算子不同的输出做不同的AIPP处理，如果该参数没有配置，默认对related_input_rank指定的模型输入的所有输出边做AIPP。 + // 配置值 <= Data算子输出边的个数。 + repeated uint32 input_edge_idx = 3; + + // [Begin] 动态AIPP参数，配置静态AIPP时无效 + uint32 max_src_image_size = 4; + + // 是否支持旋转。默认不支持，开启支持旋转时，会有额外的空间和性能损失 + bool support_rotation = 5; + + // [End] 动态AIPP参数 + + + // [Begin] 静态AIPP参数，配置动态AIPP时无效 + InputFormat input_format = 51; + bool csc_switch = 52; + float cpadding_value = 53; + bool rbuv_swap_switch = 54; + bool ax_swap_switch = 55; + bool single_line_mode = 56; + + int32 src_image_size_w = 57; + int32 src_image_size_h = 58; + + bool crop = 59; + int32 load_start_pos_w = 60; + int32 load_start_pos_h = 61; + int32 crop_size_w = 62; + int32 crop_size_h = 63; + + bool resize = 64; + int32 resize_output_w = 65; + int32 resize_output_h = 66; + + bool padding = 67; + int32 left_padding_size = 68; + int32 right_padding_size = 69; + int32 top_padding_size = 70; + int32 bottom_padding_size = 71; + + int32 mean_chn_0 = 10; + int32 mean_chn_1 = 11; + int32 mean_chn_2 = 12; + int32 mean_chn_3 = 19; + float min_chn_0 = 13; + float min_chn_1 = 14; + float min_chn_2 = 15; + float min_chn_3 = 20; + repeated float var_reci_chn_0 = 16; + repeated float var_reci_chn_1 = 17; + repeated float var_reci_chn_2 = 18; + repeated float var_reci_chn_3 = 21; + + repeated int32 matrix_r0c0 = 30; + repeated int32 matrix_r0c1 = 31; + repeated int32 matrix_r0c2 = 32; + repeated int32 matrix_r1c0 = 33; + repeated int32 matrix_r1c1 = 34; + repeated int32 matrix_r1c2 = 35; + repeated int32 matrix_r2c0 = 36; + repeated int32 matrix_r2c1 = 37; + repeated int32 matrix_r2c2 = 38; + repeated int32 output_bias_0 = 39; + repeated int32 output_bias_1 = 40; + repeated int32 output_bias_2 = 41; + repeated int32 input_bias_0 = 42; + repeated int32 input_bias_1 = 43; + repeated int32 input_bias_2 = 44; + + // [End] 静态AIPP参数 + + // The n number that is used for raw/rgbir data into f16 transformation. + // The transformation equation is x/(2^n). If set to 0, no transform is performed. + uint32 raw_rgbir_to_f16_n = 45; +} + +message MultiShapeOpParams { + enum MultiShapeMode { + batch = 0; //动态batch + resolution = 1; //动态分辨率，扩展用 + } + + MultiShapeMode mode = 1; //算子模式 + uint32 related_input_rank = 2; //新增算子插入到哪个输入 + + + repeated uint32 batch_list = 11; //batch_list值，batch_list的个数是2到8之间 +} diff --git a/ge/executor/proto/om.proto b/ge/executor/proto/om.proto new file mode 100644 index 00000000..e15e5f80 --- /dev/null +++ b/ge/executor/proto/om.proto @@ -0,0 +1,396 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +package domi; + +enum TargetType +{ + MINI = 0; + TINY = 1; + LITE = 2; +} + +// offline model +message ModelDef { + string name = 1; + uint32 version = 2; + + uint64 memory_size = 10; + uint32 stream_num = 11; + uint32 event_num = 12; + uint64 weight_size = 13; + uint32 label_num = 15; + repeated OpDef op = 20; + TargetType target_type = 23; + + map attr = 30; +}; + +// operator define +message OpDef { + string name = 1; + string type = 2; + + uint32 id = 3; + uint32 stream_id = 4; + + repeated string input_name = 5; + + repeated string src_name = 8; + repeated int32 src_index = 9; + repeated int64 input = 10; + repeated int64 output = 11; + repeated TensorDescriptor input_desc = 12; + repeated TensorDescriptor output_desc = 13; + repeated WeightDef weights = 14; + repeated string dst_name = 15; + repeated int32 dst_index = 16; + + repeated int64 workspace = 20; + repeated uint32 workspace_bytes = 21; + + repeated string weight_name = 22; + repeated bool is_input_const = 23; + + map attr = 30; + + QuantizeFactorParams quantize_factor = 31; + + oneof op_params { + // start at 100 here + SendOpParams sender_param = 100; + RecvOpParams receiver_param = 200; + ConvolutionOpParams convolution_param = 300; + PoolingOpParams pooling_param = 400; + EltwiseOpParams eltwise_param = 500; + BatchNormOpParams batchnorm_param = 600; + ScaleOpParams scale_param = 700; + FullConnectionOpParams full_connection_param = 800; + SoftmaxOpParams softmax_param = 900; + ActivationOpParams activation_param = 1000; + ReshapeOpParams reshape_param = 1100; + } +}; + +message SendOpParams { + uint32 event_id = 1; +}; + +message RecvOpParams { + uint32 event_id = 1; +}; + +enum QuantizeScaleType +{ + VECTOR_SCALE = 0; + SCALAR_SCALE = 1; +} + +enum QuantizeScaleMode +{ + NORMAL_MODE = 0; + SQRT_MODE = 1; +} + +enum QuantizeAlgorithm +{ + NON_OFFSET_ALGO = 0; + HALF_OFFSET_ALGO = 1; + ALL_OFFSET_ALGO = 2; +} +message QuantizeFactor +{ + QuantizeScaleMode scale_mode = 1; + bytes scale_value = 2; + int64 scale_offset = 3; + bytes offset_data_value = 4; + int64 offset_data_offset = 5; + bytes offset_weight_value = 6; + int64 offset_weight_offset = 7; + bytes offset_pad_value = 8; + int64 offset_pad_offset = 9; +}; + +message QuantizeCalcFactor +{ + bytes offsetw = 1; + int64 offsetw_offset = 2; + bytes offsetd = 3; + int64 offsetd_offset = 4; + bytes scalereq = 5; + int64 scaledreq_offset = 6; + bytes offsetdnext = 7; + int64 offsetdnext_offset = 8; +} + +message QuantizeFactorParams +{ + QuantizeAlgorithm quantize_algo = 1; + QuantizeScaleType scale_type = 2; + QuantizeFactor quantize_param = 3; + QuantizeFactor dequantize_param = 4; + QuantizeFactor requantize_param = 5; + QuantizeCalcFactor quantizecalc_param = 6; +}; + +message ConvolutionOpParams { + int32 mode = 1; + int32 algo = 2; + int32 pad_mode = 3; + uint32 group = 4; + uint32 num_output = 5; + + repeated uint32 pad = 10; + repeated uint32 stride = 11; + repeated uint32 dilation = 12; + repeated uint32 kernel = 13; + + float alpha = 20; + float beta = 21; + + WeightDef filter = 40; + WeightDef bias = 41; + + bool relu_flag = 62; + repeated uint32 adj = 70; + repeated uint32 target_shape = 71; + repeated uint32 before_pad = 72; +}; + +message PoolingOpParams { + int32 mode = 1; + int32 nan_opt = 2; + int32 pad_mode = 3; + bool global_pooling = 4; + + repeated uint32 window = 10; + repeated uint32 pad = 11; + repeated uint32 stride = 12; + bool ceil_mode = 13; + int32 data_mode = 14; + + float alpha = 20; + float beta = 21; + repeated uint32 before_pad = 22; +}; + +message EltwiseOpParams { + int32 mode = 1; + repeated float coeff = 2; + float alpha = 3; + float beta = 4; + repeated WeightDef weight = 5; + bool relu_flag = 6; +}; + +message ActivationOpParams { + int32 mode = 1; + float coef = 2; + float alpha = 3; + float beta = 4; +}; + +message BatchNormOpParams { + int32 mode = 1; + + float alpha = 2; + float beta = 3; + double epsilon = 4;//optinal,[default = 1e-5] + bool use_global_stats = 5; //optinal,by default true,testing mode + float moving_average_fraction = 6; //optinal,[default = .999]; + + WeightDef estimated_mean = 7; + WeightDef estimated_variance = 8; + + WeightDef scale = 9; + WeightDef bias = 10; +}; + +message ScaleOpParams { + WeightDef scale = 1; + WeightDef bias = 2; +}; + +message ReshapeOpParams { + float alpha = 1; + float beta = 2; + ShapeDef shape = 3; + int32 axis = 4; + int32 num_axes = 5; + int32 format = 6; +}; + +message SoftmaxOpParams { + int32 algo = 1; + int32 mode = 2; + float alpha = 3; + float beta = 4; +}; + +message FullConnectionOpParams { + WeightDef filter = 1; + WeightDef bias = 2; + uint32 num_output = 3; + bool relu_flag = 12; +}; + +message FlattenOpParams { + float alpha = 1; + float beta = 2; + int32 start_axis = 3; + int32 end_axis = 4; +} + +message AddLimitedOpParams { + float alpha = 1; + float beta = 2; + int32 axis = 3; + bool broadcast = 4; + + repeated WeightDef weight = 10; +}; + +message MulLimitedOpParams { + float alpha = 1; + float beta = 2; + int32 axis = 3; + bool broadcast = 4; + + repeated WeightDef weight = 10; +}; + +message AddOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message MulOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message SubOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message BiasAddOpParams { + float alpha = 1; + float beta = 2; + + WeightDef bias = 10; +}; + +message MatMulOpParams { + float alpha = 1; + float beta = 2; + bool transposeX = 3; + bool transposeW = 4; + + WeightDef filter = 10; + WeightDef bias = 12; +}; + +message RsqrtOpParams { + float alpha = 1; + float beta = 2; +}; + + +message WeightDef { + int32 format = 1; + int32 data_type = 2; + ShapeDef shape = 3; + bytes data = 4; + int64 data_offset = 5; + uint32 cmps_size = 6; + bytes cmps_tab = 7; + int64 cmps_tab_offset = 10; + CompressInfo cmps_info = 8; + AllOffsetQuantizeInfo alloffset_quantize_info = 11; +} + +message ShapeDef { + repeated int64 dim = 1; +} + +enum DeviceType { + NPU = 0; // In default, we will use NPU. + CPU = 1; // CPU +} + +message AllOffsetQuantizeInfo { + float scale = 1; + int32 offset = 2; +} + +message TensorDescriptor { + int32 format = 1; + int32 data_type = 2; + repeated int64 dim = 3; + uint32 size = 4; + bool reuse_input = 5; + bool output_tensor = 7; + DeviceType device_type = 8; + bool input_tensor = 9; + uint32 real_dim_cnt = 10; + uint32 reuse_input_index = 11; + AllOffsetQuantizeInfo alloffset_quantize_info = 12; +} + +message CompressInfo { + int32 blockRow = 1; // block row + int32 blockCol = 2; // block col + int32 fractalK = 3; // fractal K + int32 fractalN = 4; // fractal N + int32 lastFractalK = 5; // K of last fractal + int32 lastFractalN = 6; // N of last fractal + int32 cubeSize = 7; // cube's length + int32 loadDir = 8; // data load directtiono 0:col load 1:row load +} + +message AttrDef { + message ListValue { + repeated string s = 2; // "list(string)" + repeated int64 i = 3 [packed = true]; // "list(int)" + repeated float f = 4 [packed = true]; // "list(float)" + repeated bool b = 5 [packed = true]; // "list(bool)" + repeated uint32 u = 6 [packed = true]; // "list(uint)" + repeated bytes bt = 7; + } + + oneof value { + string s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + uint32 u = 6; // "uint32" + bytes bt = 7; + ListValue list = 1; // any "list(...)" + NamedAttrs func = 10; + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NamedAttrs { + string name = 1; + map attr = 2; +} + diff --git a/ge/executor/proto/op_mapping_info.proto b/ge/executor/proto/op_mapping_info.proto new file mode 100644 index 00000000..e23b7ebe --- /dev/null +++ b/ge/executor/proto/op_mapping_info.proto @@ -0,0 +1,73 @@ +syntax = "proto3"; +package aicpu.dump; + +message Shape { + repeated uint64 dim = 1; +} + +message Output { + int32 data_type = 1; + int32 format = 2; + Shape shape = 3; + uint64 address = 4; + string original_name = 5; + int32 original_output_index = 6; + int32 original_output_data_type = 7; + int32 original_output_format = 8; + uint64 size = 9; +} + +message Input { + int32 data_type =1; + int32 format = 2; + Shape shape = 3; + uint64 address = 4; + uint64 size = 5; +} + +enum BufferType { + L1 = 0; +} + +message OpBuffer { + BufferType buffer_type = 1; + uint64 address = 2; + uint64 size = 3; +} + +message Op { + string op_name = 1; + string op_type = 2; +} + +message Task { + uint32 task_id = 1; + uint32 stream_id = 2; + Op op = 3; + repeated Output output = 4; + bool end_graph = 5; + repeated Input input = 6; + repeated OpBuffer buffer = 7; +} + +message OpMappingInfo { + string dump_path = 1; + oneof model_name_param { + string model_name = 2; + } + oneof model_id_param { + uint32 model_id = 3; + } + oneof step_id { + uint64 step_id_addr = 4; + } + oneof iterations_per_loop { + uint64 iterations_per_loop_addr = 5; + } + oneof loop_cond { + uint64 loop_cond_addr = 6; + } + uint32 flag = 7; // 0x01 load, 0x00 unload + repeated Task task = 8; + string dump_step = 9; +} \ No newline at end of file diff --git a/ge/executor/proto/task.proto b/ge/executor/proto/task.proto new file mode 100644 index 00000000..d0c09840 --- /dev/null +++ b/ge/executor/proto/task.proto @@ -0,0 +1,165 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +package domi; + +message ModelTaskDef { + string version = 1; + + map attr = 9; // Extended field + repeated TaskDef task = 10; + + uint64 memory_size = 11; + uint32 stream_num = 12; + uint32 event_num = 13; + uint64 weight_size = 14; + + repeated bytes op = 15; // input/output opdef in bytes + + uint64 base_addr = 16; // base addr + uint64 weight_addr = 17; // weight addr + uint32 batch_num = 18; +} + + +message TaskDef { + uint32 id = 1; + uint32 type = 2; + + uint32 stream_id = 10; + uint32 event_id = 11; + + KernelDef kernel = 20; + KernelExDef kernel_ex = 21; + KernelHcclDef kernel_hccl = 25; + EventExDef event_ex = 26; + LogTimeStampDef log_timestamp = 28; + + uint32 label_id = 30; + + MemcpyAsyncDef memcpy_async = 31; + StreamSwitchDef stream_switch = 32; + StreamActiveDef stream_active = 33; + bytes private_def = 34; + uint64 ops_kernel_store_ptr = 35; // adjustments to other fields in the future + StreamSwitchNDef stream_switch_n = 36; + + LabelSetDef label_set = 37; + LabelGotoExDef label_goto_ex = 38; + LabelSwitchByIndexDef label_switch_by_index = 39; +} + +message KernelDef { + KernelContext context = 1; + + string stub_func = 10; + uint32 block_dim = 11; + uint32 args_size = 12; + bytes args = 13; + bytes sm_desc = 14; + bytes flowtable = 15; + string so_name = 16; + string kernel_name = 17; + bytes kernel_ext_info = 18; + uint32 kernel_ext_info_size = 19; +} + +message KernelContext { + uint32 kernel_type = 1; + uint32 op_id = 2; // OP type in CCE + uint32 kernel_func_id = 3; + uint32 op_index = 4; // TE/Custom operator + bool is_flowtable = 5; // Identify whether args is a flowtable structure + bytes args_offset = 6; // args offset information + uint32 args_count = 7; // args count + repeated uint32 origin_op_index = 8; +} + + +message KernelExDef { + uint32 flags = 1; + + uint32 op_index = 4; + uint32 args_size = 12; + bytes args = 13; + bytes task_info = 14; // serialized nodeDef, funcDef, inputoutput + uint32 task_info_size = 15; + bytes kernel_ext_info = 16; + uint32 kernel_ext_info_size = 17; +} + + +message KernelHcclDef { + uint32 op_index = 8; + string hccl_type = 9; +} + + +message EventExDef { + uint32 op_index = 1; + uint32 event_type = 2; +} + +message LogTimeStampDef { + uint64 logid = 1; + bool notify = 2; + uint32 flat = 3; +} + +message MemcpyAsyncDef { + uint64 dst = 1; + uint64 dst_max = 2; + uint64 src = 3; + uint64 count = 4; + uint32 kind = 5; + uint32 op_index = 6; +} + +message StreamSwitchDef { + uint32 op_index = 1; + uint32 true_stream_id = 2; + int64 value = 3; + uint64 value_ptr = 4; + uint32 data_type = 5; +} + +message StreamActiveDef { + uint32 op_index = 1; + uint32 active_stream_id = 2; +} + +message StreamSwitchNDef { + uint32 op_index = 1; + uint32 size = 2; + repeated int64 target_value = 3; + repeated uint32 true_stream_id = 4; + uint32 element_size = 5; + uint32 data_type = 6; +} + +message LabelSetDef { + uint32 op_index = 1; + uint32 label_id = 2; + uint32 model_id = 3; +} + +message LabelGotoExDef { + uint32 op_index = 1; + uint32 label_id = 2; + uint32 model_id = 3; +} + +message LabelSwitchByIndexDef { + uint32 op_index = 1; + uint32 label_max = 2; +} diff --git a/ge/offline/proto/ge_ir.proto b/ge/offline/proto/ge_ir.proto new file mode 100644 index 00000000..e7bfe0cb --- /dev/null +++ b/ge/offline/proto/ge_ir.proto @@ -0,0 +1,190 @@ +syntax = "proto3"; + +package ge.proto; + +enum DataType +{ + DT_UNDEFINED = 0; // Used to indicate a DataType field has not been set. + DT_FLOAT = 1; // float type + DT_FLOAT16 = 2; // fp16 type + DT_INT8 = 3; // int8 type + DT_UINT8 = 4; // uint8 type + DT_INT16 = 5; // int16 type + DT_UINT16 = 6; // uint16 type + DT_INT32 = 7; // + DT_INT64 = 8; // int64 type + DT_UINT32 = 9; // unsigned int32 + DT_UINT64 = 10; // unsigned int64 + DT_BOOL = 11; // bool type + DT_DOUBLE = 12; // double type + DT_STRING = 13; // string type + DT_DUAL_SUB_INT8 = 14; /**< dual output int8 type */ + DT_DUAL_SUB_UINT8 = 15; /**< dual output uint8 type */ + DT_COMPLEX64 = 16; // complex64 type + DT_COMPLEX128 = 17; // complex128 type + DT_QINT8 = 18; // qint8 type + DT_QINT16 = 19; // qint16 type + DT_QINT32 = 20; // qint32 type + DT_QUINT8 = 21; // quint8 type + DT_QUINT16 = 22; // quint16 type + DT_RESOURCE = 23; // resource type + DT_STRING_REF = 24; // string_ref type + DT_DUAL = 25; /**< dual output type */ +} + +message AttrDef +{ + message ListValue + { + enum ListValueType{ + VT_LIST_NONE = 0; + VT_LIST_STRING = 1; + VT_LIST_INT = 2; + VT_LIST_FLOAT = 3; + VT_LIST_BOOL = 4; + VT_LIST_BYTES = 5; + VT_LIST_TENSOR_DESC = 6; + VT_LIST_TENSOR = 7; + VT_LIST_GRAPH = 8; + VT_LIST_NAMED_ATTRS = 9; + VT_LIST_DATA_TYPE = 10; + } + repeated bytes s = 2; // "list(string)" + repeated int64 i = 3; // "list(int)" + repeated float f = 4; // "list(float)" + repeated bool b = 5; // "list(bool)" + repeated bytes bt = 7; + repeated TensorDescriptor td = 8; + repeated TensorDef t = 9; + repeated GraphDef g = 10; + repeated NamedAttrs na = 11; + repeated int64 dt = 12; // list ge::DataType + + ListValueType val_type = 20; + } + + message ListListInt{ + message ListInt{ + repeated int64 list_i = 1; // list int + } + repeated ListInt list_list_i = 1; // list list int + } + + oneof value + { + bytes s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + bytes bt = 7; + ListValue list = 1; // any "list(...)" + NamedAttrs func = 10; // Used to support attr nesting + TensorDescriptor td = 11; // GeTensorDesc type + TensorDef t = 12; // GeTensor type + GraphDef g = 13; // Graph type + ListListInt list_list_int = 14; // List List Int type + int64 dt = 15; // ge::DataType + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NamedAttrs +{ + string name = 1; + map attr = 2; +} + +// Shape / dimension description, using row-major order +message ShapeDef +{ + repeated int64 dim = 1; // Size of each dimension +} + +// Multidimensional data description +message TensorDescriptor +{ + string name = 1; // Optional parameter, tensor name + + DataType dtype = 2; // tensor datatype + ShapeDef shape = 3; // Shape / dimension + string layout = 4; // Tensor format, eg: "NCHW", "NHWC", "CHW", "ND" + + bool has_out_attr = 9; + int64 size = 10; + int64 weight_size = 11; + bool reuse_input = 12; + bool output_tensor = 13; + string device_type = 14; + bool input_tensor =15; + int64 real_dim_cnt = 16; + int64 reuse_input_index = 17; + int64 data_offset = 18; + int64 cmps_size = 19; + string cmps_tab = 20; + int64 cmps_tab_offset = 21; + + map attr = 5; // Set of extra parameter fields +} + +// GeTensor definition +message TensorDef +{ + TensorDescriptor desc = 1; // Tensor description + bytes data = 2; // Tensor data +} + + +// Operator description +message OpDef +{ + string name = 1; // name + string type = 2; // type + + repeated string input = 5; // input original op name + outgoing index. op_name锛歩ndex + + map attr = 10; // Set of operator parameter fields + + bool has_out_attr = 20; + int64 id = 21; + int64 stream_id =22; + repeated string input_name = 23; + repeated string src_name = 24; + repeated int64 src_index = 25; + repeated string dst_name = 26; + repeated int64 dst_index = 27; + repeated int64 input_i = 28; + repeated int64 output_i = 29; + repeated int64 workspace = 30; + repeated int64 workspace_bytes = 31; + repeated bool is_input_const = 32; + repeated TensorDescriptor input_desc = 33; + repeated TensorDescriptor output_desc = 34; + repeated string subgraph_name = 35; +} + +// Graph definition +message GraphDef +{ + string name = 1; // name + + repeated string input = 4; // Graph input + repeated string output = 5; // Graph output + + repeated OpDef op = 6; // List of operators + + map attr = 11; // Extended field +} + +// model definition +message ModelDef +{ + string name = 1; // name + uint32 version = 2; // IR Proto verion + string custom_version = 3; // User model version number, passed in by user + + repeated GraphDef graph = 7; // Graph definition锛実raph[0] represents the main diagram in modeldef + + map attr = 11; // Extended field +} + diff --git a/ge/offline/proto/insert_op.proto b/ge/offline/proto/insert_op.proto new file mode 100644 index 00000000..c635ca14 --- /dev/null +++ b/ge/offline/proto/insert_op.proto @@ -0,0 +1,136 @@ +syntax = "proto3"; + +package domi; + +message InsertNewOps { + repeated AippOpParams aipp_op = 1; + repeated MultiShapeOpParams multi_shape_op = 2; +} + +message AippOpParams { + enum InputFormat { + UNDEFINED = 0; + YUV420SP_U8 = 1; + XRGB8888_U8 = 2; + RGB888_U8 = 3; + YUV400_U8 = 4; + NC1HWC0DI_FP16 = 5; + NC1HWC0DI_S8 = 6; + ARGB8888_U8 = 7; + YUYV_U8 = 8; + YUV422SP_U8 = 9; + AYUV444_U8 = 10; + RAW10 = 11; + RAW12 = 12; + RAW16 = 13; + RAW24 = 14; + RGB16 = 15; + RGB20 = 16; + RGB24 = 17; + RGB8_IR = 18; + RGB16_IR = 19; + RGB24_IR = 20; + } + + enum AippMode { + undefined = 0; + static = 1; + dynamic = 2; + } + + // AIPP模式，区分静态AIPP和动态AIPP + AippMode aipp_mode = 1; + + // related_input_rank参数为必填，类型为整型，配置范围>=0, <=输入Data算子的个数，默认值为0。 + // 标识对模型的第几个输入做AIPP处理，例如模型有两个输入，需要对第2个输入做AIPP，则配置related_input_rank为1。 + uint32 related_input_rank = 2; + + // input_edge_idx参数为可选，类型为整型，配置范围为>=0。 + // 配置该参数的作用，在于对Data算子不同的输出做不同的AIPP处理，如果该参数没有配置，默认对related_input_rank指定的模型输入的所有输出边做AIPP。 + // 配置值 <= Data算子输出边的个数。 + repeated uint32 input_edge_idx = 3; + + // [Begin] 动态AIPP参数，配置静态AIPP时无效 + uint32 max_src_image_size = 4; + + // 是否支持旋转。默认不支持，开启支持旋转时，会有额外的空间和性能损失 + bool support_rotation = 5; + + // [End] 动态AIPP参数 + + + // [Begin] 静态AIPP参数，配置动态AIPP时无效 + InputFormat input_format = 51; + bool csc_switch = 52; + float cpadding_value = 53; + bool rbuv_swap_switch = 54; + bool ax_swap_switch = 55; + bool single_line_mode = 56; + + int32 src_image_size_w = 57; + int32 src_image_size_h = 58; + + bool crop = 59; + int32 load_start_pos_w = 60; + int32 load_start_pos_h = 61; + int32 crop_size_w = 62; + int32 crop_size_h = 63; + + bool resize = 64; + int32 resize_output_w = 65; + int32 resize_output_h = 66; + + bool padding = 67; + int32 left_padding_size = 68; + int32 right_padding_size = 69; + int32 top_padding_size = 70; + int32 bottom_padding_size = 71; + + int32 mean_chn_0 = 10; + int32 mean_chn_1 = 11; + int32 mean_chn_2 = 12; + int32 mean_chn_3 = 19; + float min_chn_0 = 13; + float min_chn_1 = 14; + float min_chn_2 = 15; + float min_chn_3 = 20; + repeated float var_reci_chn_0 = 16; + repeated float var_reci_chn_1 = 17; + repeated float var_reci_chn_2 = 18; + repeated float var_reci_chn_3 = 21; + + repeated int32 matrix_r0c0 = 30; + repeated int32 matrix_r0c1 = 31; + repeated int32 matrix_r0c2 = 32; + repeated int32 matrix_r1c0 = 33; + repeated int32 matrix_r1c1 = 34; + repeated int32 matrix_r1c2 = 35; + repeated int32 matrix_r2c0 = 36; + repeated int32 matrix_r2c1 = 37; + repeated int32 matrix_r2c2 = 38; + repeated int32 output_bias_0 = 39; + repeated int32 output_bias_1 = 40; + repeated int32 output_bias_2 = 41; + repeated int32 input_bias_0 = 42; + repeated int32 input_bias_1 = 43; + repeated int32 input_bias_2 = 44; + + // [End] 静态AIPP参数 + + // The n number that is used for raw/rgbir data into f16 transformation. + // The transformation equation is x/(2^n). If set to 0, no transform is performed. + uint32 raw_rgbir_to_f16_n = 45; +} + +message MultiShapeOpParams { + enum MultiShapeMode { + batch = 0; //动态batch + resolution = 1; //动态分辨率，扩展用 + } + + MultiShapeMode mode = 1; //算子模式 + uint32 related_input_rank = 2; //新增算子插入到哪个输入 + + + repeated uint32 batch_list = 11; //batch_list值，batch_list的个数是2到8之间 +} diff --git a/ge/offline/proto/om.proto b/ge/offline/proto/om.proto new file mode 100644 index 00000000..e15e5f80 --- /dev/null +++ b/ge/offline/proto/om.proto @@ -0,0 +1,396 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +package domi; + +enum TargetType +{ + MINI = 0; + TINY = 1; + LITE = 2; +} + +// offline model +message ModelDef { + string name = 1; + uint32 version = 2; + + uint64 memory_size = 10; + uint32 stream_num = 11; + uint32 event_num = 12; + uint64 weight_size = 13; + uint32 label_num = 15; + repeated OpDef op = 20; + TargetType target_type = 23; + + map attr = 30; +}; + +// operator define +message OpDef { + string name = 1; + string type = 2; + + uint32 id = 3; + uint32 stream_id = 4; + + repeated string input_name = 5; + + repeated string src_name = 8; + repeated int32 src_index = 9; + repeated int64 input = 10; + repeated int64 output = 11; + repeated TensorDescriptor input_desc = 12; + repeated TensorDescriptor output_desc = 13; + repeated WeightDef weights = 14; + repeated string dst_name = 15; + repeated int32 dst_index = 16; + + repeated int64 workspace = 20; + repeated uint32 workspace_bytes = 21; + + repeated string weight_name = 22; + repeated bool is_input_const = 23; + + map attr = 30; + + QuantizeFactorParams quantize_factor = 31; + + oneof op_params { + // start at 100 here + SendOpParams sender_param = 100; + RecvOpParams receiver_param = 200; + ConvolutionOpParams convolution_param = 300; + PoolingOpParams pooling_param = 400; + EltwiseOpParams eltwise_param = 500; + BatchNormOpParams batchnorm_param = 600; + ScaleOpParams scale_param = 700; + FullConnectionOpParams full_connection_param = 800; + SoftmaxOpParams softmax_param = 900; + ActivationOpParams activation_param = 1000; + ReshapeOpParams reshape_param = 1100; + } +}; + +message SendOpParams { + uint32 event_id = 1; +}; + +message RecvOpParams { + uint32 event_id = 1; +}; + +enum QuantizeScaleType +{ + VECTOR_SCALE = 0; + SCALAR_SCALE = 1; +} + +enum QuantizeScaleMode +{ + NORMAL_MODE = 0; + SQRT_MODE = 1; +} + +enum QuantizeAlgorithm +{ + NON_OFFSET_ALGO = 0; + HALF_OFFSET_ALGO = 1; + ALL_OFFSET_ALGO = 2; +} +message QuantizeFactor +{ + QuantizeScaleMode scale_mode = 1; + bytes scale_value = 2; + int64 scale_offset = 3; + bytes offset_data_value = 4; + int64 offset_data_offset = 5; + bytes offset_weight_value = 6; + int64 offset_weight_offset = 7; + bytes offset_pad_value = 8; + int64 offset_pad_offset = 9; +}; + +message QuantizeCalcFactor +{ + bytes offsetw = 1; + int64 offsetw_offset = 2; + bytes offsetd = 3; + int64 offsetd_offset = 4; + bytes scalereq = 5; + int64 scaledreq_offset = 6; + bytes offsetdnext = 7; + int64 offsetdnext_offset = 8; +} + +message QuantizeFactorParams +{ + QuantizeAlgorithm quantize_algo = 1; + QuantizeScaleType scale_type = 2; + QuantizeFactor quantize_param = 3; + QuantizeFactor dequantize_param = 4; + QuantizeFactor requantize_param = 5; + QuantizeCalcFactor quantizecalc_param = 6; +}; + +message ConvolutionOpParams { + int32 mode = 1; + int32 algo = 2; + int32 pad_mode = 3; + uint32 group = 4; + uint32 num_output = 5; + + repeated uint32 pad = 10; + repeated uint32 stride = 11; + repeated uint32 dilation = 12; + repeated uint32 kernel = 13; + + float alpha = 20; + float beta = 21; + + WeightDef filter = 40; + WeightDef bias = 41; + + bool relu_flag = 62; + repeated uint32 adj = 70; + repeated uint32 target_shape = 71; + repeated uint32 before_pad = 72; +}; + +message PoolingOpParams { + int32 mode = 1; + int32 nan_opt = 2; + int32 pad_mode = 3; + bool global_pooling = 4; + + repeated uint32 window = 10; + repeated uint32 pad = 11; + repeated uint32 stride = 12; + bool ceil_mode = 13; + int32 data_mode = 14; + + float alpha = 20; + float beta = 21; + repeated uint32 before_pad = 22; +}; + +message EltwiseOpParams { + int32 mode = 1; + repeated float coeff = 2; + float alpha = 3; + float beta = 4; + repeated WeightDef weight = 5; + bool relu_flag = 6; +}; + +message ActivationOpParams { + int32 mode = 1; + float coef = 2; + float alpha = 3; + float beta = 4; +}; + +message BatchNormOpParams { + int32 mode = 1; + + float alpha = 2; + float beta = 3; + double epsilon = 4;//optinal,[default = 1e-5] + bool use_global_stats = 5; //optinal,by default true,testing mode + float moving_average_fraction = 6; //optinal,[default = .999]; + + WeightDef estimated_mean = 7; + WeightDef estimated_variance = 8; + + WeightDef scale = 9; + WeightDef bias = 10; +}; + +message ScaleOpParams { + WeightDef scale = 1; + WeightDef bias = 2; +}; + +message ReshapeOpParams { + float alpha = 1; + float beta = 2; + ShapeDef shape = 3; + int32 axis = 4; + int32 num_axes = 5; + int32 format = 6; +}; + +message SoftmaxOpParams { + int32 algo = 1; + int32 mode = 2; + float alpha = 3; + float beta = 4; +}; + +message FullConnectionOpParams { + WeightDef filter = 1; + WeightDef bias = 2; + uint32 num_output = 3; + bool relu_flag = 12; +}; + +message FlattenOpParams { + float alpha = 1; + float beta = 2; + int32 start_axis = 3; + int32 end_axis = 4; +} + +message AddLimitedOpParams { + float alpha = 1; + float beta = 2; + int32 axis = 3; + bool broadcast = 4; + + repeated WeightDef weight = 10; +}; + +message MulLimitedOpParams { + float alpha = 1; + float beta = 2; + int32 axis = 3; + bool broadcast = 4; + + repeated WeightDef weight = 10; +}; + +message AddOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message MulOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message SubOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message BiasAddOpParams { + float alpha = 1; + float beta = 2; + + WeightDef bias = 10; +}; + +message MatMulOpParams { + float alpha = 1; + float beta = 2; + bool transposeX = 3; + bool transposeW = 4; + + WeightDef filter = 10; + WeightDef bias = 12; +}; + +message RsqrtOpParams { + float alpha = 1; + float beta = 2; +}; + + +message WeightDef { + int32 format = 1; + int32 data_type = 2; + ShapeDef shape = 3; + bytes data = 4; + int64 data_offset = 5; + uint32 cmps_size = 6; + bytes cmps_tab = 7; + int64 cmps_tab_offset = 10; + CompressInfo cmps_info = 8; + AllOffsetQuantizeInfo alloffset_quantize_info = 11; +} + +message ShapeDef { + repeated int64 dim = 1; +} + +enum DeviceType { + NPU = 0; // In default, we will use NPU. + CPU = 1; // CPU +} + +message AllOffsetQuantizeInfo { + float scale = 1; + int32 offset = 2; +} + +message TensorDescriptor { + int32 format = 1; + int32 data_type = 2; + repeated int64 dim = 3; + uint32 size = 4; + bool reuse_input = 5; + bool output_tensor = 7; + DeviceType device_type = 8; + bool input_tensor = 9; + uint32 real_dim_cnt = 10; + uint32 reuse_input_index = 11; + AllOffsetQuantizeInfo alloffset_quantize_info = 12; +} + +message CompressInfo { + int32 blockRow = 1; // block row + int32 blockCol = 2; // block col + int32 fractalK = 3; // fractal K + int32 fractalN = 4; // fractal N + int32 lastFractalK = 5; // K of last fractal + int32 lastFractalN = 6; // N of last fractal + int32 cubeSize = 7; // cube's length + int32 loadDir = 8; // data load directtiono 0:col load 1:row load +} + +message AttrDef { + message ListValue { + repeated string s = 2; // "list(string)" + repeated int64 i = 3 [packed = true]; // "list(int)" + repeated float f = 4 [packed = true]; // "list(float)" + repeated bool b = 5 [packed = true]; // "list(bool)" + repeated uint32 u = 6 [packed = true]; // "list(uint)" + repeated bytes bt = 7; + } + + oneof value { + string s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + uint32 u = 6; // "uint32" + bytes bt = 7; + ListValue list = 1; // any "list(...)" + NamedAttrs func = 10; + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NamedAttrs { + string name = 1; + map attr = 2; +} + diff --git a/ge/offline/proto/task.proto b/ge/offline/proto/task.proto new file mode 100644 index 00000000..d0c09840 --- /dev/null +++ b/ge/offline/proto/task.proto @@ -0,0 +1,165 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +package domi; + +message ModelTaskDef { + string version = 1; + + map attr = 9; // Extended field + repeated TaskDef task = 10; + + uint64 memory_size = 11; + uint32 stream_num = 12; + uint32 event_num = 13; + uint64 weight_size = 14; + + repeated bytes op = 15; // input/output opdef in bytes + + uint64 base_addr = 16; // base addr + uint64 weight_addr = 17; // weight addr + uint32 batch_num = 18; +} + + +message TaskDef { + uint32 id = 1; + uint32 type = 2; + + uint32 stream_id = 10; + uint32 event_id = 11; + + KernelDef kernel = 20; + KernelExDef kernel_ex = 21; + KernelHcclDef kernel_hccl = 25; + EventExDef event_ex = 26; + LogTimeStampDef log_timestamp = 28; + + uint32 label_id = 30; + + MemcpyAsyncDef memcpy_async = 31; + StreamSwitchDef stream_switch = 32; + StreamActiveDef stream_active = 33; + bytes private_def = 34; + uint64 ops_kernel_store_ptr = 35; // adjustments to other fields in the future + StreamSwitchNDef stream_switch_n = 36; + + LabelSetDef label_set = 37; + LabelGotoExDef label_goto_ex = 38; + LabelSwitchByIndexDef label_switch_by_index = 39; +} + +message KernelDef { + KernelContext context = 1; + + string stub_func = 10; + uint32 block_dim = 11; + uint32 args_size = 12; + bytes args = 13; + bytes sm_desc = 14; + bytes flowtable = 15; + string so_name = 16; + string kernel_name = 17; + bytes kernel_ext_info = 18; + uint32 kernel_ext_info_size = 19; +} + +message KernelContext { + uint32 kernel_type = 1; + uint32 op_id = 2; // OP type in CCE + uint32 kernel_func_id = 3; + uint32 op_index = 4; // TE/Custom operator + bool is_flowtable = 5; // Identify whether args is a flowtable structure + bytes args_offset = 6; // args offset information + uint32 args_count = 7; // args count + repeated uint32 origin_op_index = 8; +} + + +message KernelExDef { + uint32 flags = 1; + + uint32 op_index = 4; + uint32 args_size = 12; + bytes args = 13; + bytes task_info = 14; // serialized nodeDef, funcDef, inputoutput + uint32 task_info_size = 15; + bytes kernel_ext_info = 16; + uint32 kernel_ext_info_size = 17; +} + + +message KernelHcclDef { + uint32 op_index = 8; + string hccl_type = 9; +} + + +message EventExDef { + uint32 op_index = 1; + uint32 event_type = 2; +} + +message LogTimeStampDef { + uint64 logid = 1; + bool notify = 2; + uint32 flat = 3; +} + +message MemcpyAsyncDef { + uint64 dst = 1; + uint64 dst_max = 2; + uint64 src = 3; + uint64 count = 4; + uint32 kind = 5; + uint32 op_index = 6; +} + +message StreamSwitchDef { + uint32 op_index = 1; + uint32 true_stream_id = 2; + int64 value = 3; + uint64 value_ptr = 4; + uint32 data_type = 5; +} + +message StreamActiveDef { + uint32 op_index = 1; + uint32 active_stream_id = 2; +} + +message StreamSwitchNDef { + uint32 op_index = 1; + uint32 size = 2; + repeated int64 target_value = 3; + repeated uint32 true_stream_id = 4; + uint32 element_size = 5; + uint32 data_type = 6; +} + +message LabelSetDef { + uint32 op_index = 1; + uint32 label_id = 2; + uint32 model_id = 3; +} + +message LabelGotoExDef { + uint32 op_index = 1; + uint32 label_id = 2; + uint32 model_id = 3; +} + +message LabelSwitchByIndexDef { + uint32 op_index = 1; + uint32 label_max = 2; +} diff --git a/ge/proto/caffe/caffe.proto b/ge/proto/caffe/caffe.proto new file mode 100644 index 00000000..3f45aae2 --- /dev/null +++ b/ge/proto/caffe/caffe.proto @@ -0,0 +1,1821 @@ +syntax = "proto2"; + +package domi.caffe; + +// Specifies the shape (dimensions) of a Blob. +message BlobShape { + repeated int64 dim = 1 [packed = true]; +} + +message BlobProto { + optional BlobShape shape = 7; + repeated float data = 5 [packed = true]; + repeated float diff = 6 [packed = true]; + repeated double double_data = 8 [packed = true]; + repeated double double_diff = 9 [packed = true]; + optional bytes int8_data = 10; + repeated int32 int32_data = 11 [packed = true]; + repeated uint64 uint64_data = 12 [packed = true]; + // 4D dimensions -- deprecated. Use "shape" instead. + optional int32 num = 1 [default = 0]; + optional int32 channels = 2 [default = 0]; + optional int32 height = 3 [default = 0]; + optional int32 width = 4 [default = 0]; +} + +// The BlobProtoVector is simply a way to pass multiple blobproto instances +// around. +message BlobProtoVector { + repeated BlobProto blobs = 1; +} + +message Datum { + optional int32 channels = 1; + optional int32 height = 2; + optional int32 width = 3; + // the actual image data, in bytes + optional bytes data = 4; + optional int32 label = 5; + // Optionally, the datum could also hold float data. + repeated float float_data = 6; + // If true data contains an encoded image that need to be decoded + optional bool encoded = 7 [default = false]; +} + +message FillerParameter { + // The filler type. + optional string type = 1 [default = 'constant']; + optional float value = 2 [default = 0]; // the value in constant filler + optional float min = 3 [default = 0]; // the min value in uniform filler + optional float max = 4 [default = 1]; // the max value in uniform filler + optional float mean = 5 [default = 0]; // the mean value in Gaussian filler + optional float std = 6 [default = 1]; // the std value in Gaussian filler + // The expected number of non-zero output weights for a given input in + // Gaussian filler -- the default -1 means don't perform sparsification. + optional int32 sparse = 7 [default = -1]; + // Normalize the filler variance by fan_in, fan_out, or their average. + // Applies to 'xavier' and 'msra' fillers. + enum VarianceNorm { + FAN_IN = 0; + FAN_OUT = 1; + AVERAGE = 2; + } + optional VarianceNorm variance_norm = 8 [default = FAN_IN]; +} + +message NetParameter { + optional string name = 1; // consider giving the network a name + // DEPRECATED. See InputParameter. The input blobs to the network. + repeated string input = 3; + // DEPRECATED. See InputParameter. The shape of the input blobs. + repeated BlobShape input_shape = 8; + + // 4D input dimensions -- deprecated. Use "input_shape" instead. + // If specified, for each input blob there should be four + // values specifying the num, channels, height and width of the input blob. + // Thus, there should be a total of (4 * #input) numbers. + repeated int32 input_dim = 4; + + // Whether the network will force every layer to carry out backward operation. + // If set False, then whether to carry out backward is determined + // automatically according to the net structure and learning rates. + optional bool force_backward = 5 [default = false]; + // The current "state" of the network, including the phase, level, and stage. + // Some layers may be included/excluded depending on this state and the states + // specified in the layers' include and exclude fields. + optional NetState state = 6; + + // Print debugging information about results while running Net::Forward, + // Net::Backward, and Net::Update. + optional bool debug_info = 7 [default = false]; + + // The layers that make up the net. Each of their configurations, including + // connectivity and behavior, is specified as a LayerParameter. + repeated LayerParameter layer = 100; // ID 100 so layers are printed last. + + // DEPRECATED: use 'layer' instead. + repeated V1LayerParameter layers = 2; +} + +// NOTE +// Update the next available ID when you add a new SolverParameter field. +// +// SolverParameter next available ID: 42 (last added: layer_wise_reduce) +message SolverParameter { + ////////////////////////////////////////////////////////////////////////////// + // Specifying the train and test networks + // + // Exactly one train net must be specified using one of the following fields: + // train_net_param, train_net, net_param, net + // One or more test nets may be specified using any of the following fields: + // test_net_param, test_net, net_param, net + // If more than one test net field is specified (e.g., both net and + // test_net are specified), they will be evaluated in the field order given + // above: (1) test_net_param, (2) test_net, (3) net_param/net. + // A test_iter must be specified for each test_net. + // A test_level and/or a test_stage may also be specified for each test_net. + ////////////////////////////////////////////////////////////////////////////// + + // Proto filename for the train net, possibly combined with one or more + // test nets. + optional string net = 24; + // Inline train net param, possibly combined with one or more test nets. + optional NetParameter net_param = 25; + + optional string train_net = 1; // Proto filename for the train net. + repeated string test_net = 2; // Proto filenames for the test nets. + optional NetParameter train_net_param = 21; // Inline train net params. + repeated NetParameter test_net_param = 22; // Inline test net params. + + // The states for the train/test nets. Must be unspecified or + // specified once per net. + // + // By default, all states will have solver = true; + // train_state will have phase = TRAIN, + // and all test_state's will have phase = TEST. + // Other defaults are set according to the NetState defaults. + optional NetState train_state = 26; + repeated NetState test_state = 27; + + // The number of iterations for each test net. + repeated int32 test_iter = 3; + + // The number of iterations between two testing phases. + optional int32 test_interval = 4 [default = 0]; + optional bool test_compute_loss = 19 [default = false]; + // If true, run an initial test pass before the first iteration, + // ensuring memory availability and printing the starting value of the loss. + optional bool test_initialization = 32 [default = true]; + optional float base_lr = 5; // The base learning rate + // the number of iterations between displaying info. If display = 0, no info + // will be displayed. + optional int32 display = 6; + // Display the loss averaged over the last average_loss iterations + optional int32 average_loss = 33 [default = 1]; + optional int32 max_iter = 7; // the maximum number of iterations + // accumulate gradients over `iter_size` x `batch_size` instances + optional int32 iter_size = 36 [default = 1]; + + // The learning rate decay policy. The currently implemented learning rate + // policies are as follows: + // - fixed: always return base_lr. + // - step: return base_lr * gamma ^ (floor(iter / step)) + // - exp: return base_lr * gamma ^ iter + // - inv: return base_lr * (1 + gamma * iter) ^ (- power) + // - multistep: similar to step but it allows non uniform steps defined by + // stepvalue + // - poly: the effective learning rate follows a polynomial decay, to be + // zero by the max_iter. return base_lr (1 - iter/max_iter) ^ (power) + // - sigmoid: the effective learning rate follows a sigmod decay + // return base_lr ( 1/(1 + exp(-gamma * (iter - stepsize)))) + // + // where base_lr, max_iter, gamma, step, stepvalue and power are defined + // in the solver parameter protocol buffer, and iter is the current iteration. + optional string lr_policy = 8; + optional float gamma = 9; // The parameter to compute the learning rate. + optional float power = 10; // The parameter to compute the learning rate. + optional float momentum = 11; // The momentum value. + optional float weight_decay = 12; // The weight decay. + // regularization types supported: L1 and L2 + // controlled by weight_decay + optional string regularization_type = 29 [default = "L2"]; + // the stepsize for learning rate policy "step" + optional int32 stepsize = 13; + // the stepsize for learning rate policy "multistep" + repeated int32 stepvalue = 34; + + // Set clip_gradients to >= 0 to clip parameter gradients to that L2 norm, + // whenever their actual L2 norm is larger. + optional float clip_gradients = 35 [default = -1]; + + optional int32 snapshot = 14 [default = 0]; // The snapshot interval + optional string snapshot_prefix = 15; // The prefix for the snapshot. + // whether to snapshot diff in the results or not. Snapshotting diff will help + // debugging but the final protocol buffer size will be much larger. + optional bool snapshot_diff = 16 [default = false]; + enum SnapshotFormat { + HDF5 = 0; + BINARYPROTO = 1; + } + optional SnapshotFormat snapshot_format = 37 [default = BINARYPROTO]; + // the mode solver will use: 0 for CPU and 1 for GPU. Use GPU in default. + enum SolverMode { + CPU = 0; + GPU = 1; + } + optional SolverMode solver_mode = 17 [default = GPU]; + // the device_id will that be used in GPU mode. Use device_id = 0 in default. + optional int32 device_id = 18 [default = 0]; + // If non-negative, the seed with which the Solver will initialize the Caffe + // random number generator -- useful for reproducible results. Otherwise, + // (and by default) initialize using a seed derived from the system clock. + optional int64 random_seed = 20 [default = -1]; + + // type of the solver + optional string type = 40 [default = "SGD"]; + + // numerical stability for RMSProp, AdaGrad and AdaDelta and Adam + optional float delta = 31 [default = 1e-8]; + // parameters for the Adam solver + optional float momentum2 = 39 [default = 0.999]; + + // RMSProp decay value + // MeanSquare(t) = rms_decay*MeanSquare(t-1) + (1-rms_decay)*SquareGradient(t) + optional float rms_decay = 38 [default = 0.99]; + + // If true, print information about the state of the net that may help with + // debugging learning problems. + optional bool debug_info = 23 [default = false]; + + // If false, don't save a snapshot after training finishes. + optional bool snapshot_after_train = 28 [default = true]; + + // DEPRECATED: old solver enum types, use string instead + enum SolverType { + SGD = 0; + NESTEROV = 1; + ADAGRAD = 2; + RMSPROP = 3; + ADADELTA = 4; + ADAM = 5; + } + // DEPRECATED: use type instead of solver_type + optional SolverType solver_type = 30 [default = SGD]; + + // Overlap compute and communication for data parallel training + optional bool layer_wise_reduce = 41 [default = true]; +} + +// A message that stores the solver snapshots +message SolverState { + optional int32 iter = 1; // The current iteration + optional string learned_net = 2; // The file that stores the learned net. + repeated BlobProto history = 3; // The history for sgd solvers + optional int32 current_step = 4 [default = 0]; // The current step for learning rate +} + +enum Phase { + TRAIN = 0; + TEST = 1; +} + +message NetState { + optional Phase phase = 1 [default = TEST]; + optional int32 level = 2 [default = 0]; + repeated string stage = 3; +} + +message NetStateRule { + // Set phase to require the NetState have a particular phase (TRAIN or TEST) + // to meet this rule. + optional Phase phase = 1; + + // Set the minimum and/or maximum levels in which the layer should be used. + // Leave undefined to meet the rule regardless of level. + optional int32 min_level = 2; + optional int32 max_level = 3; + + // Customizable sets of stages to include or exclude. + // The net must have ALL of the specified stages and NONE of the specified + // "not_stage"s to meet the rule. + // (Use multiple NetStateRules to specify conjunctions of stages.) + repeated string stage = 4; + repeated string not_stage = 5; +} + +// Specifies training parameters (multipliers on global learning constants, +// and the name and other settings used for weight sharing). +message ParamSpec { + // The names of the parameter blobs -- useful for sharing parameters among + // layers, but never required otherwise. To share a parameter between two + // layers, give it a (non-empty) name. + optional string name = 1; + + // Whether to require shared weights to have the same shape, or just the same + // count -- defaults to STRICT if unspecified. + optional DimCheckMode share_mode = 2; + enum DimCheckMode { + // STRICT (default) requires that num, channels, height, width each match. + STRICT = 0; + // PERMISSIVE requires only the count (num*channels*height*width) to match. + PERMISSIVE = 1; + } + + // The multiplier on the global learning rate for this parameter. + optional float lr_mult = 3 [default = 1.0]; + + // The multiplier on the global weight decay for this parameter. + optional float decay_mult = 4 [default = 1.0]; +} + +// NOTE +// Update the next available ID when you add a new LayerParameter field. +// +// LayerParameter next available layer-specific ID: 151 (last added: smooth_l1_loss_param) +message LayerParameter { + optional string name = 1; // the layer name + optional string type = 2; // the layer type + repeated string bottom = 3; // the name of each bottom blob + repeated string top = 4; // the name of each top blob + + // The train / test phase for computation. + optional Phase phase = 10; + + // The amount of weight to assign each top blob in the objective. + // Each layer assigns a default value, usually of either 0 or 1, + // to each top blob. + repeated float loss_weight = 5; + + // Specifies training parameters (multipliers on global learning constants, + // and the name and other settings used for weight sharing). + repeated ParamSpec param = 6; + + // The blobs containing the numeric parameters of the layer. + repeated BlobProto blobs = 7; + + // Specifies whether to backpropagate to each bottom. If unspecified, + // Caffe will automatically infer whether each input needs backpropagation + // to compute parameter gradients. If set to true for some inputs, + // backpropagation to those inputs is forced; if set false for some inputs, + // backpropagation to those inputs is skipped. + // + // The size must be either 0 or equal to the number of bottoms. + repeated bool propagate_down = 11; + + // Rules controlling whether and when a layer is included in the network, + // based on the current NetState. You may specify a non-zero number of rules + // to include OR exclude, but not both. If no include or exclude rules are + // specified, the layer is always included. If the current NetState meets + // ANY (i.e., one or more) of the specified rules, the layer is + // included/excluded. + repeated NetStateRule include = 8; + repeated NetStateRule exclude = 9; + + // Parameters for data pre-processing. + optional TransformationParameter transform_param = 100; + + // Parameters shared by loss layers. + optional LossParameter loss_param = 101; + + // Layer type-specific parameters. + // + // Note: certain layers may have more than one computational engine + // for their implementation. These layers include an Engine type and + // engine parameter for selecting the implementation. + // The default for the engine is set by the ENGINE switch at compile-time. + optional AccuracyParameter accuracy_param = 102; + optional ArgMaxParameter argmax_param = 103; + optional BatchNormParameter batch_norm_param = 139; + optional BiasParameter bias_param = 141; + optional ConcatParameter concat_param = 104; + optional ContrastiveLossParameter contrastive_loss_param = 105; + optional ConvolutionParameter convolution_param = 106; + optional CropParameter crop_param = 144; + optional DataParameter data_param = 107; + optional DetectionOutputParameter detection_output_param = 150; + optional DropoutParameter dropout_param = 108; + optional DummyDataParameter dummy_data_param = 109; + optional EltwiseParameter eltwise_param = 110; + optional ELUParameter elu_param = 140; + optional EmbedParameter embed_param = 137; + optional ExpParameter exp_param = 111; + optional FlattenParameter flatten_param = 135; + optional HDF5DataParameter hdf5_data_param = 112; + optional HDF5OutputParameter hdf5_output_param = 113; + optional HingeLossParameter hinge_loss_param = 114; + optional ImageDataParameter image_data_param = 115; + optional InfogainLossParameter infogain_loss_param = 116; + optional InnerProductParameter inner_product_param = 117; + optional InputParameter input_param = 143; + optional LogParameter log_param = 134; + optional LRNParameter lrn_param = 118; + optional MemoryDataParameter memory_data_param = 119; + optional MVNParameter mvn_param = 120; + optional ParameterParameter parameter_param = 145; + optional PoolingParameter pooling_param = 121; + optional PowerParameter power_param = 122; + optional PReLUParameter prelu_param = 131; + optional PythonParameter python_param = 130; + optional RecurrentParameter recurrent_param = 146; + optional ReductionParameter reduction_param = 136; + optional ReLUParameter relu_param = 123; + optional ReshapeParameter reshape_param = 133; + optional ScaleParameter scale_param = 142; + optional SigmoidParameter sigmoid_param = 124; + optional SmoothL1LossParameter smooth_l1_loss_param = 148; + optional SoftmaxParameter softmax_param = 125; + optional SPPParameter spp_param = 132; + optional SliceParameter slice_param = 126; + optional TanHParameter tanh_param = 127; + optional ThresholdParameter threshold_param = 128; + optional TileParameter tile_param = 138; + optional WindowDataParameter window_data_param = 129; + optional PermuteParameter permute_param = 202; + optional PriorBoxParameter prior_box_param = 203; + optional NormalizeParameter norm_param = 206; + optional PSROIPoolingParameter psroi_pooling_param = 207; + optional FreespaceExtractParameter freespace_extract_param = 151; + optional PostprocessParameter postprocess_param = 152; + optional SpatialTransformParameter spatial_transform_param = 153; + optional ROIAlignParameter roi_align_param = 154; + optional ReorgParameter reorg_param = 155; + optional RegionParameter region_param = 156; + optional ReverseParameter reverse_param = 157; + optional InterpParameter interp_param = 158; + optional ShuffleChannelParameter shuffle_channel_param = 159; + optional UpsampleParameter upsample_param = 160; + optional ROIPoolingParameter roi_pooling_param = 161; + optional YoloParameter yolo_param = 199; + optional YoloV3DetectionOutputParameter yolov3_detection_output_param = 200; + optional ProposalParameter proposal_param = 201; + optional FSRDetectionOutputParameter fsrdetectionoutput_param = 222; + optional SSDDetectionOutputParameter ssddetectionoutput_param = 232; + optional YoloV2DetectionOutputParameter yolov2_detection_output_param = 204; + optional QuantParameter quant_param = 208; + optional CondTakeParameter condtake_param = 233; + optional MatrixInverseParameter matrix_inverse_param = 210; + optional WarpPerspectiveParameter warp_perspective_param = 234; + optional BatchMatMulParameter batch_matmul_param = 235; + optional SpatialTransformerParameter st_param = 5000; + optional YoloV3DetectionOutputV2Parameter yolov3_detection_output_v2_param = 5001; +} + +// Message that stores parameters used to apply transformation +// to the data layer's data +message TransformationParameter { + // For data pre-processing, we can do simple scaling and subtracting the + // data mean, if provided. Note that the mean subtraction is always carried + // out before scaling. + optional float scale = 1 [default = 1]; + // Specify if we want to randomly mirror data. + optional bool mirror = 2 [default = false]; + // Specify if we would like to randomly crop an image. + optional uint32 crop_size = 3 [default = 0]; + // mean_file and mean_value cannot be specified at the same time + optional string mean_file = 4; + // if specified can be repeated once (would substract it from all the channels) + // or can be repeated the same number of times as channels + // (would subtract them from the corresponding channel) + repeated float mean_value = 5; + // Force the decoded image to have 3 color channels. + optional bool force_color = 6 [default = false]; + // Force the decoded image to have 1 color channels. + optional bool force_gray = 7 [default = false]; +} + +// Message that stores parameters shared by loss layers +message LossParameter { + // If specified, ignore instances with the given label. + optional int32 ignore_label = 1; + // How to normalize the loss for loss layers that aggregate across batches, + // spatial dimensions, or other dimensions. Currently only implemented in + // SoftmaxWithLoss and SigmoidCrossEntropyLoss layers. + enum NormalizationMode { + // Divide by the number of examples in the batch times spatial dimensions. + // Outputs that receive the ignore label will NOT be ignored in computing + // the normalization factor. + FULL = 0; + // Divide by the total number of output locations that do not take the + // ignore_label. If ignore_label is not set, this behaves like FULL. + VALID = 1; + // Divide by the batch size. + BATCH_SIZE = 2; + // Do not normalize the loss. + NONE = 3; + } + // For historical reasons, the default normalization for + // SigmoidCrossEntropyLoss is BATCH_SIZE and *not* VALID. + optional NormalizationMode normalization = 3 [default = VALID]; + // Deprecated. Ignored if normalization is specified. If normalization + // is not specified, then setting this to false will be equivalent to + // normalization = BATCH_SIZE to be consistent with previous behavior. + optional bool normalize = 2; +} + +// Messages that store parameters used by individual layer types follow, in +// alphabetical order. + +message AccuracyParameter { + // When computing accuracy, count as correct by comparing the true label to + // the top k scoring classes. By default, only compare to the top scoring + // class (i.e. argmax). + optional uint32 top_k = 1 [default = 1]; + + // The "label" axis of the prediction blob, whose argmax corresponds to the + // predicted label -- may be negative to index from the end (e.g., -1 for the + // last axis). For example, if axis == 1 and the predictions are + // (N x C x H x W), the label blob is expected to contain N*H*W ground truth + // labels with integer values in {0, 1, ..., C-1}. + optional int32 axis = 2 [default = 1]; + + // If specified, ignore instances with the given label. + optional int32 ignore_label = 3; +} + +message ArgMaxParameter { + // If true produce pairs (argmax, maxval) + optional bool out_max_val = 1 [default = false]; + optional uint32 top_k = 2 [default = 1]; + // The axis along which to maximise -- may be negative to index from the + // end (e.g., -1 for the last axis). + // By default ArgMaxLayer maximizes over the flattened trailing dimensions + // for each index of the first / num dimension. + optional int32 axis = 3; +} + +message ConcatParameter { + // The axis along which to concatenate -- may be negative to index from the + // end (e.g., -1 for the last axis). Other axes must have the + // same dimension for all the bottom blobs. + // By default, ConcatLayer concatenates blobs along the "channels" axis (1). + optional int32 axis = 2 [default = 1]; + + // DEPRECATED: alias for "axis" -- does not support negative indexing. + optional uint32 concat_dim = 1 [default = 1]; +} + +message BatchNormParameter { + // If false, normalization is performed over the current mini-batch + // and global statistics are accumulated (but not yet used) by a moving + // average. + // If true, those accumulated mean and variance values are used for the + // normalization. + // By default, it is set to false when the network is in the training + // phase and true when the network is in the testing phase. + optional bool use_global_stats = 1; + // What fraction of the moving average remains each iteration? + // Smaller values make the moving average decay faster, giving more + // weight to the recent values. + // Each iteration updates the moving average @f$S_{t-1}@f$ with the + // current mean @f$ Y_t @f$ by + // @f$ S_t = (1-\beta)Y_t + \beta \cdot S_{t-1} @f$, where @f$ \beta @f$ + // is the moving_average_fraction parameter. + optional float moving_average_fraction = 2 [default = .999]; + // Small value to add to the variance estimate so that we don't divide by + // zero. + optional float eps = 3 [default = 1e-5]; +} + +message BiasParameter { + // The first axis of bottom[0] (the first input Blob) along which to apply + // bottom[1] (the second input Blob). May be negative to index from the end + // (e.g., -1 for the last axis). + // + // For example, if bottom[0] is 4D with shape 100x3x40x60, the output + // top[0] will have the same shape, and bottom[1] may have any of the + // following shapes (for the given value of axis): + // (axis == 0 == -4) 100; 100x3; 100x3x40; 100x3x40x60 + // (axis == 1 == -3) 3; 3x40; 3x40x60 + // (axis == 2 == -2) 40; 40x60 + // (axis == 3 == -1) 60 + // Furthermore, bottom[1] may have the empty shape (regardless of the value of + // "axis") -- a scalar bias. + optional int32 axis = 1 [default = 1]; + + // (num_axes is ignored unless just one bottom is given and the bias is + // a learned parameter of the layer. Otherwise, num_axes is determined by the + // number of axes by the second bottom.) + // The number of axes of the input (bottom[0]) covered by the bias + // parameter, or -1 to cover all axes of bottom[0] starting from `axis`. + // Set num_axes := 0, to add a zero-axis Blob: a scalar. + optional int32 num_axes = 2 [default = 1]; + + // (filler is ignored unless just one bottom is given and the bias is + // a learned parameter of the layer.) + // The initialization for the learned bias parameter. + // Default is the zero (0) initialization, resulting in the BiasLayer + // initially performing the identity operation. + optional FillerParameter filler = 3; + optional bool bias_from_blob = 4 [default = true]; +} + +message ContrastiveLossParameter { + // margin for dissimilar pair + optional float margin = 1 [default = 1.0]; + // The first implementation of this cost did not exactly match the cost of + // Hadsell et al 2006 -- using (margin - d^2) instead of (margin - d)^2. + // legacy_version = false (the default) uses (margin - d)^2 as proposed in the + // Hadsell paper. New models should probably use this version. + // legacy_version = true uses (margin - d^2). This is kept to support / + // reproduce existing models and results + optional bool legacy_version = 2 [default = false]; +} + +message ConvolutionParameter { + optional uint32 num_output = 1; // The number of outputs for the layer + optional bool bias_term = 2 [default = true]; // whether to have bias terms + + // Pad, kernel size, and stride are all given as a single value for equal + // dimensions in all spatial dimensions, or once per spatial dimension. + repeated uint32 pad = 3; // The padding size; defaults to 0 + repeated uint32 kernel_size = 4; // The kernel size + repeated uint32 stride = 6; // The stride; defaults to 1 + // Factor used to dilate the kernel, (implicitly) zero-filling the resulting + // holes. (Kernel dilation is sometimes referred to by its use in the + // algorithme 脿 trous from Holschneider et al. 1987.) + repeated uint32 dilation = 18; // The dilation; defaults to 1 + + // For 2D convolution only, the *_h and *_w versions may also be used to + // specify both spatial dimensions. + optional uint32 pad_h = 9 [default = 0]; // The padding height (2D only) + optional uint32 pad_w = 10 [default = 0]; // The padding width (2D only) + optional uint32 kernel_h = 11; // The kernel height (2D only) + optional uint32 kernel_w = 12; // The kernel width (2D only) + optional uint32 stride_h = 13; // The stride height (2D only) + optional uint32 stride_w = 14; // The stride width (2D only) + + optional uint32 group = 5 [default = 1]; // The group size for group conv + + optional FillerParameter weight_filler = 7; // The filler for the weight + optional FillerParameter bias_filler = 8; // The filler for the bias + enum Engine { + DEFAULT = 0; + CAFFE = 1; + CUDNN = 2; + } + optional Engine engine = 15 [default = DEFAULT]; + + // The axis to interpret as "channels" when performing convolution. + // Preceding dimensions are treated as independent inputs; + // succeeding dimensions are treated as "spatial". + // With (N, C, H, W) inputs, and axis == 1 (the default), we perform + // N independent 2D convolutions, sliding C-channel (or (C/g)-channels, for + // groups g>1) filters across the spatial axes (H, W) of the input. + // With (N, C, D, H, W) inputs, and axis == 1, we perform + // N independent 3D convolutions, sliding (C/g)-channels + // filters across the spatial axes (D, H, W) of the input. + optional int32 axis = 16 [default = 1]; + + // Whether to force use of the general ND convolution, even if a specific + // implementation for blobs of the appropriate number of spatial dimensions + // is available. (Currently, there is only a 2D-specific convolution + // implementation; for input blobs with num_axes != 2, this option is + // ignored and the ND implementation will be used.) + optional bool force_nd_im2col = 17 [default = false]; +} + +message CropParameter { + // To crop, elements of the first bottom are selected to fit the dimensions + // of the second, reference bottom. The crop is configured by + // - the crop `axis` to pick the dimensions for cropping + // - the crop `offset` to set the shift for all/each dimension + // to align the cropped bottom with the reference bottom. + // All dimensions up to but excluding `axis` are preserved, while + // the dimensions including and trailing `axis` are cropped. + // If only one `offset` is set, then all dimensions are offset by this amount. + // Otherwise, the number of offsets must equal the number of cropped axes to + // shift the crop in each dimension accordingly. + // Note: standard dimensions are N,C,H,W so the default is a spatial crop, + // and `axis` may be negative to index from the end (e.g., -1 for the last + // axis). + optional int32 axis = 1 [default = 2]; + repeated uint32 offset = 2; +} + +message DataParameter { + enum DB { + LEVELDB = 0; + LMDB = 1; + } + // Specify the data source. + optional string source = 1; + // Specify the batch size. + optional uint32 batch_size = 4; + // The rand_skip variable is for the data layer to skip a few data points + // to avoid all asynchronous sgd clients to start at the same point. The skip + // point would be set as rand_skip * rand(0,1). Note that rand_skip should not + // be larger than the number of keys in the database. + // DEPRECATED. Each solver accesses a different subset of the database. + optional uint32 rand_skip = 7 [default = 0]; + optional DB backend = 8 [default = LEVELDB]; + // DEPRECATED. See TransformationParameter. For data pre-processing, we can do + // simple scaling and subtracting the data mean, if provided. Note that the + // mean subtraction is always carried out before scaling. + optional float scale = 2 [default = 1]; + optional string mean_file = 3; + // DEPRECATED. See TransformationParameter. Specify if we would like to randomly + // crop an image. + optional uint32 crop_size = 5 [default = 0]; + // DEPRECATED. See TransformationParameter. Specify if we want to randomly mirror + // data. + optional bool mirror = 6 [default = false]; + // Force the encoded image to have 3 color channels + optional bool force_encoded_color = 9 [default = false]; + // Prefetch queue (Increase if data feeding bandwidth varies, within the + // limit of device memory for GPU training) + optional uint32 prefetch = 10 [default = 4]; +} + +message DropoutParameter { + optional float dropout_ratio = 1 [default = 0.5]; // dropout ratio + optional bool scale_train = 2 [default = true]; // scale train or test phase +} + +// DummyDataLayer fills any number of arbitrarily shaped blobs with random +// (or constant) data generated by "Fillers" (see "message FillerParameter"). +message DummyDataParameter { + // This layer produces N >= 1 top blobs. DummyDataParameter must specify 1 or N + // shape fields, and 0, 1 or N data_fillers. + // + // If 0 data_fillers are specified, ConstantFiller with a value of 0 is used. + // If 1 data_filler is specified, it is applied to all top blobs. If N are + // specified, the ith is applied to the ith top blob. + repeated FillerParameter data_filler = 1; + repeated BlobShape shape = 6; + + // 4D dimensions -- deprecated. Use "shape" instead. + repeated uint32 num = 2; + repeated uint32 channels = 3; + repeated uint32 height = 4; + repeated uint32 width = 5; +} + +message EltwiseParameter { + enum EltwiseOp { + PROD = 0; + SUM = 1; + MAX = 2; + } + optional EltwiseOp operation = 1 [default = SUM]; // element-wise operation + repeated float coeff = 2; // blob-wise coefficient for SUM operation + + // Whether to use an asymptotically slower (for >2 inputs) but stabler method + // of computing the gradient for the PROD operation. (No effect for SUM op.) + optional bool stable_prod_grad = 3 [default = true]; +} + +// Message that stores parameters used by ELULayer +message ELUParameter { + // Described in: + // Clevert, D.-A., Unterthiner, T., & Hochreiter, S. (2015). Fast and Accurate + // Deep Network Learning by Exponential Linear Units (ELUs). arXiv + optional float alpha = 1 [default = 1]; +} + +// Message that stores parameters used by EmbedLayer +message EmbedParameter { + optional uint32 num_output = 1; // The number of outputs for the layer + // The input is given as integers to be interpreted as one-hot + // vector indices with dimension num_input. Hence num_input should be + // 1 greater than the maximum possible input value. + optional uint32 input_dim = 2; + + optional bool bias_term = 3 [default = true]; // Whether to use a bias term + optional FillerParameter weight_filler = 4; // The filler for the weight + optional FillerParameter bias_filler = 5; // The filler for the bias + +} + +// Message that stores parameters used by ExpLayer +message ExpParameter { + // ExpLayer computes outputs y = base ^ (shift + scale * x), for base > 0. + // Or if base is set to the default (-1), base is set to e, + // so y = exp(shift + scale * x). + optional float base = 1 [default = -1.0]; + optional float scale = 2 [default = 1.0]; + optional float shift = 3 [default = 0.0]; +} + +/// Message that stores parameters used by FlattenLayer +message FlattenParameter { + // The first axis to flatten: all preceding axes are retained in the output. + // May be negative to index from the end (e.g., -1 for the last axis). + optional int32 axis = 1 [default = 1]; + + // The last axis to flatten: all following axes are retained in the output. + // May be negative to index from the end (e.g., the default -1 for the last + // axis). + optional int32 end_axis = 2 [default = -1]; +} + +// Message that stores parameters used by HDF5DataLayer +message HDF5DataParameter { + // Specify the data source. + optional string source = 1; + // Specify the batch size. + optional uint32 batch_size = 2; + + // Specify whether to shuffle the data. + // If shuffle == true, the ordering of the HDF5 files is shuffled, + // and the ordering of data within any given HDF5 file is shuffled, + // but data between different files are not interleaved; all of a file's + // data are output (in a random order) before moving onto another file. + optional bool shuffle = 3 [default = false]; +} + +message HDF5OutputParameter { + optional string file_name = 1; +} + +message HingeLossParameter { + enum Norm { + L1 = 1; + L2 = 2; + } + // Specify the Norm to use L1 or L2 + optional Norm norm = 1 [default = L1]; +} + +message ImageDataParameter { + // Specify the data source. + optional string source = 1; + // Specify the batch size. + optional uint32 batch_size = 4 [default = 1]; + // The rand_skip variable is for the data layer to skip a few data points + // to avoid all asynchronous sgd clients to start at the same point. The skip + // point would be set as rand_skip * rand(0,1). Note that rand_skip should not + // be larger than the number of keys in the database. + optional uint32 rand_skip = 7 [default = 0]; + // Whether or not ImageLayer should shuffle the list of files at every epoch. + optional bool shuffle = 8 [default = false]; + // It will also resize images if new_height or new_width are not zero. + optional uint32 new_height = 9 [default = 0]; + optional uint32 new_width = 10 [default = 0]; + // Specify if the images are color or gray + optional bool is_color = 11 [default = true]; + // DEPRECATED. See TransformationParameter. For data pre-processing, we can do + // simple scaling and subtracting the data mean, if provided. Note that the + // mean subtraction is always carried out before scaling. + optional float scale = 2 [default = 1]; + optional string mean_file = 3; + // DEPRECATED. See TransformationParameter. Specify if we would like to randomly + // crop an image. + optional uint32 crop_size = 5 [default = 0]; + // DEPRECATED. See TransformationParameter. Specify if we want to randomly mirror + // data. + optional bool mirror = 6 [default = false]; + optional string root_folder = 12 [default = ""]; +} + +message InfogainLossParameter { + // Specify the infogain matrix source. + optional string source = 1; + optional int32 axis = 2 [default = 1]; // axis of prob +} + +message InnerProductParameter { + optional uint32 num_output = 1; // The number of outputs for the layer + optional bool bias_term = 2 [default = true]; // whether to have bias terms + optional FillerParameter weight_filler = 3; // The filler for the weight + optional FillerParameter bias_filler = 4; // The filler for the bias + + // The first axis to be lumped into a single inner product computation; + // all preceding axes are retained in the output. + // May be negative to index from the end (e.g., -1 for the last axis). + optional int32 axis = 5 [default = 1]; + // Specify whether to transpose the weight matrix or not. + // If transpose == true, any operations will be performed on the transpose + // of the weight matrix. The weight matrix itself is not going to be transposed + // but rather the transfer flag of operations will be toggled accordingly. + optional bool transpose = 6 [default = false]; +} + +message InputParameter { + // This layer produces N >= 1 top blob(s) to be assigned manually. + // Define N shapes to set a shape for each top. + // Define 1 shape to set the same shape for every top. + // Define no shape to defer to reshaping manually. + repeated BlobShape shape = 1; +} + +// Message that stores parameters used by LogLayer +message LogParameter { + // LogLayer computes outputs y = log_base(shift + scale * x), for base > 0. + // Or if base is set to the default (-1), base is set to e, + // so y = ln(shift + scale * x) = log_e(shift + scale * x) + optional float base = 1 [default = -1.0]; + optional float scale = 2 [default = 1.0]; + optional float shift = 3 [default = 0.0]; +} + +// Message that stores parameters used by LRNLayer +message LRNParameter { + optional uint32 local_size = 1 [default = 5]; + optional float alpha = 2 [default = 1.]; + optional float beta = 3 [default = 0.75]; + enum NormRegion { + ACROSS_CHANNELS = 0; + WITHIN_CHANNEL = 1; + } + optional NormRegion norm_region = 4 [default = ACROSS_CHANNELS]; + optional float k = 5 [default = 1.]; + enum Engine { + DEFAULT = 0; + CAFFE = 1; + CUDNN = 2; + } + optional Engine engine = 6 [default = DEFAULT]; +} + +message MemoryDataParameter { + optional uint32 batch_size = 1; + optional uint32 channels = 2; + optional uint32 height = 3; + optional uint32 width = 4; +} + +message MVNParameter { + // This parameter can be set to false to normalize mean only + optional bool normalize_variance = 1 [default = true]; + + // This parameter can be set to true to perform DNN-like MVN + optional bool across_channels = 2 [default = false]; + + // Epsilon for not dividing by zero while normalizing variance + optional float eps = 3 [default = 1e-9]; +} + +message ParameterParameter { + optional BlobShape shape = 1; +} + +message PoolingParameter { + enum PoolMethod { + MAX = 0; + AVE = 1; + STOCHASTIC = 2; + } + optional PoolMethod pool = 1 [default = MAX]; // The pooling method + // Pad, kernel size, and stride are all given as a single value for equal + // dimensions in height and width or as Y, X pairs. + optional uint32 pad = 4 [default = 0]; // The padding size (equal in Y, X) + optional uint32 pad_h = 9 [default = 0]; // The padding height + optional uint32 pad_w = 10 [default = 0]; // The padding width + optional uint32 kernel_size = 2; // The kernel size (square) + optional uint32 kernel_h = 5; // The kernel height + optional uint32 kernel_w = 6; // The kernel width + optional uint32 stride = 3 [default = 1]; // The stride (equal in Y, X) + optional uint32 stride_h = 7; // The stride height + optional uint32 stride_w = 8; // The stride width + enum Engine { + DEFAULT = 0; + CAFFE = 1; + CUDNN = 2; + } + optional Engine engine = 11 [default = DEFAULT]; + // If global_pooling then it will pool over the size of the bottom by doing + // kernel_h = bottom->height and kernel_w = bottom->width + optional bool global_pooling = 12 [default = false]; + optional bool ceil_mode = 13 [default = true]; + // How to calculate the output size - using ceil (default) or floor rounding. + enum RoundMode { + CEIL = 0; + FLOOR = 1; + } + optional RoundMode round_mode = 14 [default = CEIL]; +} + +message PowerParameter { + // PowerLayer computes outputs y = (shift + scale * x) ^ power. + optional float power = 1 [default = 1.0]; + optional float scale = 2 [default = 1.0]; + optional float shift = 3 [default = 0.0]; +} + +message PythonParameter { + optional string module = 1; + optional string layer = 2; + // This value is set to the attribute `param_str` of the `PythonLayer` object + // in Python before calling the `setup()` method. This could be a number, + // string, dictionary in Python dict format, JSON, etc. You may parse this + // string in `setup` method and use it in `forward` and `backward`. + optional string param_str = 3 [default = '']; + // Whether this PythonLayer is shared among worker solvers during data parallelism. + // If true, each worker solver sequentially run forward from this layer. + // This value should be set true if you are using it as a data layer. + optional bool share_in_parallel = 4 [default = false]; +} + +// Message that stores parameters used by RecurrentLayer +message RecurrentParameter { + // The dimension of the output (and usually hidden state) representation -- + // must be explicitly set to non-zero. + optional uint32 num_output = 1 [default = 0]; + + optional FillerParameter weight_filler = 2; // The filler for the weight + optional FillerParameter bias_filler = 3; // The filler for the bias + + // Whether to enable displaying debug_info in the unrolled recurrent net. + optional bool debug_info = 4 [default = false]; + + // Whether to add as additional inputs (bottoms) the initial hidden state + // blobs, and add as additional outputs (tops) the final timestep hidden state + // blobs. The number of additional bottom/top blobs required depends on the + // recurrent architecture -- e.g., 1 for RNNs, 2 for LSTMs. + optional bool expose_hidden = 5 [default = false]; +} + +// Message that stores parameters used by ReductionLayer +message ReductionParameter { + enum ReductionOp { + SUM = 1; + ASUM = 2; + SUMSQ = 3; + MEAN = 4; + } + + optional ReductionOp operation = 1 [default = SUM]; // reduction operation + + // The first axis to reduce to a scalar -- may be negative to index from the + // end (e.g., -1 for the last axis). + // (Currently, only reduction along ALL "tail" axes is supported; reduction + // of axis M through N, where N < num_axes - 1, is unsupported.) + // Suppose we have an n-axis bottom Blob with shape: + // (d0, d1, d2, ..., d(m-1), dm, d(m+1), ..., d(n-1)). + // If axis == m, the output Blob will have shape + // (d0, d1, d2, ..., d(m-1)), + // and the ReductionOp operation is performed (d0 * d1 * d2 * ... * d(m-1)) + // times, each including (dm * d(m+1) * ... * d(n-1)) individual data. + // If axis == 0 (the default), the output Blob always has the empty shape + // (count 1), performing reduction across the entire input -- + // often useful for creating new loss functions. + optional int32 axis = 2 [default = 0]; + + optional float coeff = 3 [default = 1.0]; // coefficient for output +} + +// Message that stores parameters used by ReLULayer +message ReLUParameter { + // Allow non-zero slope for negative inputs to speed up optimization + // Described in: + // Maas, A. L., Hannun, A. Y., & Ng, A. Y. (2013). Rectifier nonlinearities + // improve neural network acoustic models. In ICML Workshop on Deep Learning + // for Audio, Speech, and Language Processing. + optional float negative_slope = 1 [default = 0]; + enum Engine { + DEFAULT = 0; + CAFFE = 1; + CUDNN = 2; + } + optional Engine engine = 2 [default = DEFAULT]; +} + +message ReshapeParameter { + // Specify the output dimensions. If some of the dimensions are set to 0, + // the corresponding dimension from the bottom layer is used (unchanged). + // Exactly one dimension may be set to -1, in which case its value is + // inferred from the count of the bottom blob and the remaining dimensions. + // For example, suppose we want to reshape a 2D blob "input" with shape 2 x 8: + // + // layer { + // type: "Reshape" bottom: "input" top: "output" + // reshape_param { ... } + // } + // + // If "input" is 2D with shape 2 x 8, then the following reshape_param + // specifications are all equivalent, producing a 3D blob "output" with shape + // 2 x 2 x 4: + // + // reshape_param { shape { dim: 2 dim: 2 dim: 4 } } + // reshape_param { shape { dim: 0 dim: 2 dim: 4 } } + // reshape_param { shape { dim: 0 dim: 2 dim: -1 } } + // reshape_param { shape { dim: 0 dim:-1 dim: 4 } } + // + optional BlobShape shape = 1; + + // axis and num_axes control the portion of the bottom blob's shape that are + // replaced by (included in) the reshape. By default (axis == 0 and + // num_axes == -1), the entire bottom blob shape is included in the reshape, + // and hence the shape field must specify the entire output shape. + // + // axis may be non-zero to retain some portion of the beginning of the input + // shape (and may be negative to index from the end; e.g., -1 to begin the + // reshape after the last axis, including nothing in the reshape, + // -2 to include only the last axis, etc.). + // + // For example, suppose "input" is a 2D blob with shape 2 x 8. + // Then the following ReshapeLayer specifications are all equivalent, + // producing a blob "output" with shape 2 x 2 x 4: + // + // reshape_param { shape { dim: 2 dim: 2 dim: 4 } } + // reshape_param { shape { dim: 2 dim: 4 } axis: 1 } + // reshape_param { shape { dim: 2 dim: 4 } axis: -3 } + // + // num_axes specifies the extent of the reshape. + // If num_axes >= 0 (and axis >= 0), the reshape will be performed only on + // input axes in the range [axis, axis+num_axes]. + // num_axes may also be -1, the default, to include all remaining axes + // (starting from axis). + // + // For example, suppose "input" is a 2D blob with shape 2 x 8. + // Then the following ReshapeLayer specifications are equivalent, + // producing a blob "output" with shape 1 x 2 x 8. + // + // reshape_param { shape { dim: 1 dim: 2 dim: 8 } } + // reshape_param { shape { dim: 1 dim: 2 } num_axes: 1 } + // reshape_param { shape { dim: 1 } num_axes: 0 } + // + // On the other hand, these would produce output blob shape 2 x 1 x 8: + // + // reshape_param { shape { dim: 2 dim: 1 dim: 8 } } + // reshape_param { shape { dim: 1 } axis: 1 num_axes: 0 } + // + optional int32 axis = 2 [default = 0]; + optional int32 num_axes = 3 [default = -1]; +} + + +message ScaleParameter { + // The first axis of bottom[0] (the first input Blob) along which to apply + // bottom[1] (the second input Blob). May be negative to index from the end + // (e.g., -1 for the last axis). + // + // For example, if bottom[0] is 4D with shape 100x3x40x60, the output + // top[0] will have the same shape, and bottom[1] may have any of the + // following shapes (for the given value of axis): + // (axis == 0 == -4) 100; 100x3; 100x3x40; 100x3x40x60 + // (axis == 1 == -3) 3; 3x40; 3x40x60 + // (axis == 2 == -2) 40; 40x60 + // (axis == 3 == -1) 60 + // Furthermore, bottom[1] may have the empty shape (regardless of the value of + // "axis") -- a scalar multiplier. + optional int32 axis = 1 [default = 1]; + + // (num_axes is ignored unless just one bottom is given and the scale is + // a learned parameter of the layer. Otherwise, num_axes is determined by the + // number of axes by the second bottom.) + // The number of axes of the input (bottom[0]) covered by the scale + // parameter, or -1 to cover all axes of bottom[0] starting from `axis`. + // Set num_axes := 0, to multiply with a zero-axis Blob: a scalar. + optional int32 num_axes = 2 [default = 1]; + + // (filler is ignored unless just one bottom is given and the scale is + // a learned parameter of the layer.) + // The initialization for the learned scale parameter. + // Default is the unit (1) initialization, resulting in the ScaleLayer + // initially performing the identity operation. + optional FillerParameter filler = 3; + + // Whether to also learn a bias (equivalent to a ScaleLayer+BiasLayer, but + // may be more efficient). Initialized with bias_filler (defaults to 0). + optional bool bias_term = 4 [default = false]; + optional FillerParameter bias_filler = 5; + optional bool scale_from_blob = 6 [default = true]; +} + +message SigmoidParameter { + enum Engine { + DEFAULT = 0; + CAFFE = 1; + CUDNN = 2; + } + optional Engine engine = 1 [default = DEFAULT]; +} + +message SliceParameter { + // The axis along which to slice -- may be negative to index from the end + // (e.g., -1 for the last axis). + // By default, SliceLayer concatenates blobs along the "channels" axis (1). + optional int32 axis = 3 [default = 1]; + repeated uint32 slice_point = 2; + + // DEPRECATED: alias for "axis" -- does not support negative indexing. + optional uint32 slice_dim = 1 [default = 1]; +} + +message SmoothL1LossParameter { + // SmoothL1Loss(x) = + // 0.5 * (sigma * x) ** 2 -- if x < 1.0 / sigma / sigma + // |x| - 0.5 / sigma / sigma -- otherwise + optional float sigma = 1 [default = 1]; +} + +// Message that stores parameters used by SoftmaxLayer, SoftmaxWithLossLayer +message SoftmaxParameter { + enum Engine { + DEFAULT = 0; + CAFFE = 1; + CUDNN = 2; + } + optional Engine engine = 1 [default = DEFAULT]; + + // The axis along which to perform the softmax -- may be negative to index + // from the end (e.g., -1 for the last axis). + // Any other axes will be evaluated as independent softmaxes. + optional int32 axis = 2 [default = 1]; +} + +message TanHParameter { + enum Engine { + DEFAULT = 0; + CAFFE = 1; + CUDNN = 2; + } + optional Engine engine = 1 [default = DEFAULT]; +} + +// Message that stores parameters used by TileLayer +message TileParameter { + // The index of the axis to tile. + optional int32 axis = 1 [default = 1]; + + // The number of copies (tiles) of the blob to output. + optional int32 tiles = 2; +} + +// Message that stores parameters used by ThresholdLayer +message ThresholdParameter { + optional float threshold = 1 [default = 0]; // Strictly positive values +} + +message WindowDataParameter { + // Specify the data source. + optional string source = 1; + // For data pre-processing, we can do simple scaling and subtracting the + // data mean, if provided. Note that the mean subtraction is always carried + // out before scaling. + optional float scale = 2 [default = 1]; + optional string mean_file = 3; + // Specify the batch size. + optional uint32 batch_size = 4; + // Specify if we would like to randomly crop an image. + optional uint32 crop_size = 5 [default = 0]; + // Specify if we want to randomly mirror data. + optional bool mirror = 6 [default = false]; + // Foreground (object) overlap threshold + optional float fg_threshold = 7 [default = 0.5]; + // Background (non-object) overlap threshold + optional float bg_threshold = 8 [default = 0.5]; + // Fraction of batch that should be foreground objects + optional float fg_fraction = 9 [default = 0.25]; + // Amount of contextual padding to add around a window + // (used only by the window_data_layer) + optional uint32 context_pad = 10 [default = 0]; + // Mode for cropping out a detection window + // warp: cropped window is warped to a fixed size and aspect ratio + // square: the tightest square around the window is cropped + optional string crop_mode = 11 [default = "warp"]; + // cache_images: will load all images in memory for faster access + optional bool cache_images = 12 [default = false]; + // append root_folder to locate images + optional string root_folder = 13 [default = ""]; +} + +message SPPParameter { + enum PoolMethod { + MAX = 0; + AVE = 1; + STOCHASTIC = 2; + } + optional uint32 pyramid_height = 1; + optional PoolMethod pool = 2 [default = MAX]; // The pooling method + enum Engine { + DEFAULT = 0; + CAFFE = 1; + CUDNN = 2; + } + optional Engine engine = 6 [default = DEFAULT]; +} + +// DEPRECATED: use LayerParameter. +message V1LayerParameter { + repeated string bottom = 2; + repeated string top = 3; + optional string name = 4; + repeated NetStateRule include = 32; + repeated NetStateRule exclude = 33; + enum LayerType { + NONE = 0; + ABSVAL = 35; + ACCURACY = 1; + ARGMAX = 30; + BNLL = 2; + CONCAT = 3; + CONTRASTIVE_LOSS = 37; + CONVOLUTION = 4; + DATA = 5; + DECONVOLUTION = 39; + DROPOUT = 6; + DUMMY_DATA = 32; + EUCLIDEAN_LOSS = 7; + ELTWISE = 25; + EXP = 38; + FLATTEN = 8; + HDF5_DATA = 9; + HDF5_OUTPUT = 10; + HINGE_LOSS = 28; + IM2COL = 11; + IMAGE_DATA = 12; + INFOGAIN_LOSS = 13; + INNER_PRODUCT = 14; + LRN = 15; + MEMORY_DATA = 29; + MULTINOMIAL_LOGISTIC_LOSS = 16; + MVN = 34; + POOLING = 17; + POWER = 26; + RELU = 18; + SIGMOID = 19; + SIGMOID_CROSS_ENTROPY_LOSS = 27; + SILENCE = 36; + SOFTMAX = 20; + SOFTMAX_LOSS = 21; + SPLIT = 22; + SLICE = 33; + TANH = 23; + WINDOW_DATA = 24; + THRESHOLD = 31; + QUANT = 208; + DEQUANT = 209; + } + optional LayerType type = 5; + repeated BlobProto blobs = 6; + repeated string param = 1001; + repeated DimCheckMode blob_share_mode = 1002; + enum DimCheckMode { + STRICT = 0; + PERMISSIVE = 1; + } + repeated float blobs_lr = 7; + repeated float weight_decay = 8; + repeated float loss_weight = 35; + optional AccuracyParameter accuracy_param = 27; + optional ArgMaxParameter argmax_param = 23; + optional ConcatParameter concat_param = 9; + optional ContrastiveLossParameter contrastive_loss_param = 40; + optional ConvolutionParameter convolution_param = 10; + optional DataParameter data_param = 11; + optional DropoutParameter dropout_param = 12; + optional DummyDataParameter dummy_data_param = 26; + optional EltwiseParameter eltwise_param = 24; + optional ExpParameter exp_param = 41; + optional HDF5DataParameter hdf5_data_param = 13; + optional HDF5OutputParameter hdf5_output_param = 14; + optional HingeLossParameter hinge_loss_param = 29; + optional ImageDataParameter image_data_param = 15; + optional InfogainLossParameter infogain_loss_param = 16; + optional InnerProductParameter inner_product_param = 17; + optional LRNParameter lrn_param = 18; + optional MemoryDataParameter memory_data_param = 22; + optional MVNParameter mvn_param = 34; + optional PoolingParameter pooling_param = 19; + optional PowerParameter power_param = 21; + optional ReLUParameter relu_param = 30; + optional SigmoidParameter sigmoid_param = 38; + optional SoftmaxParameter softmax_param = 39; + optional SliceParameter slice_param = 31; + optional TanHParameter tanh_param = 37; + optional ThresholdParameter threshold_param = 25; + optional WindowDataParameter window_data_param = 20; + optional TransformationParameter transform_param = 36; + optional LossParameter loss_param = 42; + optional V0LayerParameter layer = 1; +} + +// DEPRECATED: V0LayerParameter is the old way of specifying layer parameters +// in Caffe. We keep this message type around for legacy support. +message V0LayerParameter { + optional string name = 1; // the layer name + optional string type = 2; // the string to specify the layer type + + // Parameters to specify layers with inner products. + optional uint32 num_output = 3; // The number of outputs for the layer + optional bool biasterm = 4 [default = true]; // whether to have bias terms + optional FillerParameter weight_filler = 5; // The filler for the weight + optional FillerParameter bias_filler = 6; // The filler for the bias + + optional uint32 pad = 7 [default = 0]; // The padding size + optional uint32 kernelsize = 8; // The kernel size + optional uint32 group = 9 [default = 1]; // The group size for group conv + optional uint32 stride = 10 [default = 1]; // The stride + enum PoolMethod { + MAX = 0; + AVE = 1; + STOCHASTIC = 2; + } + optional PoolMethod pool = 11 [default = MAX]; // The pooling method + optional float dropout_ratio = 12 [default = 0.5]; // dropout ratio + + optional uint32 local_size = 13 [default = 5]; // for local response norm + optional float alpha = 14 [default = 1.]; // for local response norm + optional float beta = 15 [default = 0.75]; // for local response norm + optional float k = 22 [default = 1.]; + + // For data layers, specify the data source + optional string source = 16; + // For data pre-processing, we can do simple scaling and subtracting the + // data mean, if provided. Note that the mean subtraction is always carried + // out before scaling. + optional float scale = 17 [default = 1]; + optional string meanfile = 18; + // For data layers, specify the batch size. + optional uint32 batchsize = 19; + // For data layers, specify if we would like to randomly crop an image. + optional uint32 cropsize = 20 [default = 0]; + // For data layers, specify if we want to randomly mirror data. + optional bool mirror = 21 [default = false]; + + // The blobs containing the numeric parameters of the layer + repeated BlobProto blobs = 50; + // The ratio that is multiplied on the global learning rate. If you want to + // set the learning ratio for one blob, you need to set it for all blobs. + repeated float blobs_lr = 51; + // The weight decay that is multiplied on the global weight decay. + repeated float weight_decay = 52; + + // The rand_skip variable is for the data layer to skip a few data points + // to avoid all asynchronous sgd clients to start at the same point. The skip + // point would be set as rand_skip * rand(0,1). Note that rand_skip should not + // be larger than the number of keys in the database. + optional uint32 rand_skip = 53 [default = 0]; + + // Fields related to detection (det_*) + // foreground (object) overlap threshold + optional float det_fg_threshold = 54 [default = 0.5]; + // background (non-object) overlap threshold + optional float det_bg_threshold = 55 [default = 0.5]; + // Fraction of batch that should be foreground objects + optional float det_fg_fraction = 56 [default = 0.25]; + + // optional bool OBSOLETE_can_clobber = 57 [default = true]; + + // Amount of contextual padding to add around a window + // (used only by the window_data_layer) + optional uint32 det_context_pad = 58 [default = 0]; + + // Mode for cropping out a detection window + // warp: cropped window is warped to a fixed size and aspect ratio + // square: the tightest square around the window is cropped + optional string det_crop_mode = 59 [default = "warp"]; + + // For ReshapeLayer, one needs to specify the new dimensions. + optional int32 new_num = 60 [default = 0]; + optional int32 new_channels = 61 [default = 0]; + optional int32 new_height = 62 [default = 0]; + optional int32 new_width = 63 [default = 0]; + + // Whether or not ImageLayer should shuffle the list of files at every epoch. + // It will also resize images if new_height or new_width are not zero. + optional bool shuffle_images = 64 [default = false]; + + // For ConcatLayer, one needs to specify the dimension for concatenation, and + // the other dimensions must be the same for all the bottom blobs. + // By default it will concatenate blobs along the channels dimension. + optional uint32 concat_dim = 65 [default = 1]; + + optional HDF5OutputParameter hdf5_output_param = 1001; +} + +message PReLUParameter { + // Parametric ReLU described in K. He et al, Delving Deep into Rectifiers: + // Surpassing Human-Level Performance on ImageNet Classification, 2015. + + // Initial value of a_i. Default is a_i=0.25 for all i. + optional FillerParameter filler = 1; + // Whether or not slope parameters are shared across channels. + optional bool channel_shared = 2 [default = false]; +} + +// Message that stores parameters used by DetectionOutputLayer +//message DetectionOutputParameter { +// optional int32 num_classes = 1 [default = 21]; +// optional float nms_threshold = 2 [default = 0.3]; +// optional int32 top_k = 3; +// optional float confidence_threshold = 4 [default = 0.8]; +//} + +// Message that store parameters used by PriorBoxLayer +message PriorBoxParameter { + // Encode/decode type. + enum CodeType { + CORNER = 1; + CENTER_SIZE = 2; + CORNER_SIZE = 3; + } + // Minimum box size (in pixels). Required! + repeated float min_size = 1; + // Maximum box size (in pixels). Required! + repeated float max_size = 2; + // Various of aspect ratios. Duplicate ratios will be ignored. + // If none is provided, we use default ratio 1. + repeated float aspect_ratio = 3; + // If true, will flip each aspect ratio. + // For example, if there is aspect ratio "r", + // we will generate aspect ratio "1.0/r" as well. + optional bool flip = 4 [default = true]; + // If true, will clip the prior so that it is within [0, 1] + optional bool clip = 5 [default = false]; + // Variance for adjusting the prior bboxes. + repeated float variance = 6; + // By default, we calculate img_height, img_width, step_x, step_y based on + // bottom[0] (feat) and bottom[1] (img). Unless these values are explicitely + // provided. + // Explicitly provide the img_size. + optional uint32 img_size = 7; + // Either img_size or img_h/img_w should be specified; not both. + optional uint32 img_h = 8; + optional uint32 img_w = 9; + + // Explicitly provide the step size. + optional float step = 10; + // Either step or step_h/step_w should be specified; not both. + optional float step_h = 11; + optional float step_w = 12; + + // Offset to the top left corner of each cell. + optional float offset = 13 [default = 0.5]; +} + +// Message that stores parameters used by PermutetLayer +message PermuteParameter { + // The new orders of the axes of data. Notice it should be with + // in the same range as the input data, and it starts from 0. + // Do not provide repeated order. + repeated uint32 order = 1; +} + +message NormalizeParameter { + optional bool across_spatial = 1 [default = true]; + // Initial value of scale. Default is 1.0 for all + optional FillerParameter scale_filler = 2; + // Whether or not scale parameters are shared across channels. + optional bool channel_shared = 3 [default = true]; + // Epsilon for not dividing by zero while normalizing variance + optional float eps = 4 [default = 1e-10]; +} + +// needed by ssd +message SaveOutputParameter { + // Output directory. If not empty, we will save the results. + optional string output_directory = 1; + // Output name prefix. + optional string output_name_prefix = 2; + // Output format. + // VOC - PASCAL VOC output format. + // COCO - MS COCO output format. + optional string output_format = 3; + // If you want to output results, must also provide the following two files. + // Otherwise, we will ignore saving results. + // label map file. + optional string label_map_file = 4; + // A file which contains a list of names and sizes with same order + // of the input DB. The file is in the following format: + // name height width + // ... + optional string name_size_file = 5; + // Number of test images. It can be less than the lines specified in + // name_size_file. For example, when we only want to evaluate on part + // of the test images. + optional uint32 num_test_image = 6; + // The resize parameter used in saving the data. + // optional ResizeParameter resize_param = 7; +} + +message NonMaximumSuppressionParameter { + // Threshold to be used in nms. + optional float nms_threshold = 1 [default = 0.3]; + // Maximum number of results to be kept. + optional int32 top_k = 2; + // Parameter for adaptive nms. + optional float eta = 3 [default = 1.0]; +} + +message GeneralNmsParameter { + optional int32 post_top_k = 1 ; + optional float nms_threshold = 2 [default = 0]; + optional float iou_threshold_decay = 3 [default = 1.0]; + optional float coor_scale_factor = 4 [default = 1.0]; +} + +// Message that store parameters used by DetectionOutputLayer, ssd/fasterRcnn +message DetectionOutputParameter { + optional int32 num_classes = 1; + optional bool share_location = 2 [default = true]; + optional int32 background_label_id = 3 [default = 0]; + optional NonMaximumSuppressionParameter nms_param = 4; + optional SaveOutputParameter save_output_param = 5; + optional PriorBoxParameter.CodeType code_type = 6 [default = CENTER_SIZE]; + optional bool variance_encoded_in_target = 8 [default = true]; + optional int32 keep_top_k = 7; + optional float confidence_threshold = 9; + optional float nms_threshold = 13; + optional int32 top_k = 14; + optional int32 boxes = 15 [default = 1]; + optional bool relative = 17 [default = true]; + optional float objectness_threshold = 18 [default = 0.5]; + optional float class_threshold = 19 [default = 0.5]; + repeated float biases = 20; + optional GeneralNmsParameter general_nms_param = 21; + optional float objectness_score = 22; +} +message PSROIPoolingParameter { + required float spatial_scale = 1; + required int32 output_dim = 2; // output channel number + required int32 group_size = 3; // number of groups to encode position-sensitive score maps +} +// Message that stores parameters used by FreespaceExtractLayer +message FreespaceExtractParameter { + optional float org_height = 1; +} + +// Message that stores parameters used by DetectpostprocessLayer +message PostprocessParameter { + optional float nms_thresh = 1 [default = 0.3]; + optional float conf_thresh = 2 [default = 0.5]; + optional uint32 post_nms_topn = 3 [default = 100]; + optional uint32 cls_num = 4 [default = 12]; + repeated float bbox_reg_weights = 5; +} + +// Message that stores parameters used by SpatialTransformLayer +message SpatialTransformParameter { + optional uint32 output_h = 1 [default = 0]; + optional uint32 output_w = 2 [default = 0]; + optional float border_value = 3 [default = 0]; + repeated float affine_transform = 4; + enum Engine { + DEFAULT = 0; + CAFFE = 1; + CUDNN = 2; + } + optional Engine engine = 15 [default = DEFAULT]; +} +message ROIAlignParameter { + // Pad, kernel size, and stride are all given as a single value for equal + // dimensions in height and width or as Y, X pairs. + optional uint32 pooled_h = 1 [default = 0]; // The pooled output height + optional uint32 pooled_w = 2 [default = 0]; // The pooled output width + // Multiplicative spatial scale factor to translate ROI coords from their + // input scale to the scale used when pooling + optional float spatial_scale = 3 [default = 1]; + optional int32 sampling_ratio = 4 [default = -1]; + optional int32 roi_end_mode = 5 [default = 0]; +} + +message RegionParameter { + optional uint32 classes = 1 [default = 20]; // Category of classification + optional uint32 coords = 2 [default = 4]; // Coordinates of box + optional uint32 boxes = 3 [default = 1]; // Number of boxes predicted per grid + optional uint32 softmax = 4 [default = 0]; + optional string softmax_tree = 5 [default = ""]; + optional uint32 background = 6 [default = 0]; +} +message ReorgParameter{ + optional uint32 stride = 2 [default = 2]; + optional bool reverse = 1 [default = false]; +} +message ReverseParameter{ + repeated int32 axis = 1; +} +message InterpParameter{ + optional int32 height = 1 [default = 0];//Height of output + optional int32 width = 2 [default = 0];//Width of output + optional int32 zoom_factor = 3 [default = 1];//zoom factor + optional int32 shrink_factor = 4 [default = 1];//shrink factor + optional int32 pad_beg = 5 [default = 0];//padding at begin of input + optional int32 pad_end = 6 [default = 0];//padding at end of input +} +message ShuffleChannelParameter{ + optional uint32 group = 1[default = 1]; // The number of group +} +message UpsampleParameter{ + optional float scale = 1[default = 1]; + optional int32 stride = 2[default = 2]; + optional int32 stride_h = 3[default = 2]; + optional int32 stride_w = 4[default=2]; +} +message ROIPoolingParameter { + required int32 pooled_h = 1; + required int32 pooled_w = 2; + optional float spatial_scale = 3 [default=0.0625]; + optional float spatial_scale_h = 4; + optional float spatial_scale_w = 5; +} + +message YoloParameter { + optional int32 boxes = 1 [default = 3]; + optional int32 coords = 2 [default = 4]; + optional int32 classes = 3 [default = 80]; + optional string yolo_version = 4 [default = "V3"]; + optional bool softmax = 5 [default = false]; + optional bool background = 6 [default = false]; + optional bool softmaxtree = 7 [default = false]; +} + +message YoloV3DetectionOutputParameter { + optional int32 boxes = 1 [default = 3]; + optional int32 classes = 2 [default = 80]; + optional bool relative = 3 [default = true]; + optional float obj_threshold = 4 [default = 0.5]; + optional float score_threshold = 5 [default = 0.5]; + optional float iou_threshold = 6 [default = 0.45]; + optional int32 pre_nms_topn = 7 [default = 512]; + optional int32 post_nms_topn = 8 [default = 1024]; + repeated float biases_high = 9; + repeated float biases_mid = 10; + repeated float biases_low = 11; + optional int32 coords = 12 [default = 4]; + repeated float biases = 13; + optional bool resize_origin_img_to_net = 14 [default = false]; +} + +message YoloV3DetectionOutputV2Parameter { + optional int32 boxes = 1 [default = 3]; + optional int32 classes = 2 [default = 80]; + optional bool relative = 3 [default = true]; + optional float obj_threshold = 4 [default = 0.5]; + optional float score_threshold = 5 [default = 0.5]; + optional float iou_threshold = 6 [default = 0.45]; + optional int32 pre_nms_topn = 7 [default = 512]; + optional int32 post_nms_topn = 8 [default = 1024]; + repeated float biases_high = 9; + repeated float biases_mid = 10; + repeated float biases_low = 11; + optional int32 coords = 12 [default = 4]; + repeated float biases = 13; + optional bool resize_origin_img_to_net = 14 [default = false]; + optional int32 out_box_dim = 15 [default = 3]; +} + +message ProposalParameter { + optional float feat_stride = 1 [default = 16]; + optional float base_size = 2 [default = 16]; + optional float min_size = 3 [default = 16]; + repeated float ratio = 4; + repeated float scale = 5; + optional int32 pre_nms_topn = 6 [default = 3000]; + optional int32 post_nms_topn = 7 [default = 304]; + optional float iou_threshold = 8 [default = 0.7]; + optional bool output_actual_rois_num = 9 [default = false]; +} + +message FSRDetectionOutputParameter { + required int32 num_classes = 1; + required float score_threshold = 2; + required float iou_threshold = 3; + optional int32 batch_rois = 4 [default = 1]; +} + +message SSDDetectionOutputParameter { + required int32 num_classes= 1 [default = 2]; + optional bool share_location = 2 [default = true]; + optional int32 background_label_id = 3 [default = 0]; + optional float iou_threshold = 4 [default = 0.3]; + optional int32 top_k = 5 [default = 200]; + optional float eta = 6 [default = 1.0]; + optional bool variance_encoded_in_target = 7 [default = false]; + optional int32 code_type = 8 [default = 1]; + optional int32 keep_top_k = 9 [default = -1]; + optional float confidence_threshold = 10 [default = 0.0]; +} +message YoloV2DetectionOutputParameter { + optional int32 boxes = 1 [default = 5]; + optional int32 classes = 2 [default = 80]; + optional bool relative = 3 [default = true]; + optional float obj_threshold = 4 [default = 0.5]; + optional float score_threshold = 5 [default = 0.5]; + optional float iou_threshold = 6 [default = 0.45]; + optional int32 pre_nms_topn = 7 [default = 512]; + optional int32 post_nms_topn = 8 [default = 1024]; + repeated float biases = 9; + optional int32 coords = 10 [default = 4]; + optional bool resize_origin_img_to_net = 11 [default = false]; +} + +message QuantParameter { + optional float scale = 2; + optional bytes offset = 3; +} + +message BatchMatMulParameter{ + optional bool adj_x1 = 1 [default = false]; + optional bool adj_x2 = 2 [default = false]; +} + +message CondTakeParameter { + required string mode = 1; + required float val = 2; + optional float eps = 3 [default = 1e-06]; +} + +message MatrixInverseParameter { + optional bool adjoint = 1 [default = false]; +} + +message WarpPerspectiveParameter { + required int32 out_height = 1; + required int32 out_width = 2; + optional float constant = 3; + optional string border_type = 4 [default = 'BORDER_CONSTANT']; +} + +message SpatialTransformerParameter { + // How to use the parameter passed by localisation network + optional string transform_type = 1 [default = "affine"]; + // What is the sampling technique + optional string sampler_type = 2 [default = "bilinear"]; + + // If not set,stay same with the input dimension H and W + optional int32 output_H = 3; + optional int32 output_W = 4; + // If false, only compute dTheta, DO NOT compute dU + optional bool to_compute_dU = 5 [default = true]; + + // The default value for some parameters + optional double theta_1_1 = 6; + optional double theta_1_2 = 7; + optional double theta_1_3 = 8; + optional double theta_2_1 = 9; + optional double theta_2_2 = 10; + optional double theta_2_3 = 11; +} diff --git a/ge/proto/dump_task.proto b/ge/proto/dump_task.proto new file mode 100644 index 00000000..b1e346cd --- /dev/null +++ b/ge/proto/dump_task.proto @@ -0,0 +1,111 @@ +syntax = "proto3"; +package toolkit.dumpdata; + +enum OutputDataType { + DT_UNDEFINED = 0; + DT_FLOAT = 1; + DT_FLOAT16 = 2; + DT_INT8 = 3; + DT_UINT8 = 4; + DT_INT16 = 5; + DT_UINT16 = 6; + DT_INT32 = 7; + DT_INT64 = 8; + DT_UINT32 = 9; + DT_UINT64 = 10; + DT_BOOL = 11; + DT_DOUBLE = 12; + DT_STRING = 13; + DT_DUAL_SUB_INT8 = 14; + DT_DUAL_SUB_UINT8 = 15; + DT_COMPLEX64 = 16; + DT_COMPLEX128 = 17; + DT_QINT8 = 18; + DT_QINT16 = 19; + DT_QINT32 = 20; + DT_QUINT8 = 21; + DT_QUINT16 = 22; + DT_RESOURCE = 23; + DT_STRING_REF = 24; + DT_DUAL = 25; +} + +enum OutputFormat { + FORMAT_NCHW = 0; + FORMAT_NHWC = 1; + FORMAT_ND = 2; + FORMAT_NC1HWC0 = 3; + FORMAT_FRACTAL_Z = 4; + FORMAT_NC1C0HWPAD = 5; + FORMAT_NHWC1C0 = 6; + FORMAT_FSR_NCHW = 7; + FORMAT_FRACTAL_DECONV = 8; + FORMAT_C1HWNC0 = 9; + FORMAT_FRACTAL_DECONV_TRANSPOSE = 10; + FORMAT_FRACTAL_DECONV_SP_STRIDE_TRANS = 11; + FORMAT_NC1HWC0_C04 = 12; + FORMAT_FRACTAL_Z_C04 = 13; + FORMAT_CHWN = 14; + FORMAT_FRACTAL_DECONV_SP_STRIDE8_TRANS = 15; + FORMAT_HWCN = 16; + FORMAT_NC1KHKWHWC0 = 17; + FORMAT_BN_WEIGHT = 18; + FORMAT_FILTER_HWCK = 19; + FORMAT_HASHTABLE_LOOKUP_LOOKUPS=20; + FORMAT_HASHTABLE_LOOKUP_KEYS = 21; + FORMAT_HASHTABLE_LOOKUP_VALUE = 22; + FORMAT_HASHTABLE_LOOKUP_OUTPUT = 23; + FORMAT_HASHTABLE_LOOKUP_HITS=24; + FORMAT_C1HWNCoC0 = 25; + FORMAT_MD = 26; + FORMAT_NDHWC = 27; + FORMAT_FRACTAL_ZZ = 28; + FORMAT_FRACTAL_NZ = 29; + FORMAT_RESERVED = 30; +} + +message OriginalOp { + string name = 1; + uint32 output_index = 2; + OutputDataType data_type = 3; + OutputFormat format = 4; +} + +message Shape { + repeated uint64 dim = 1; +} + +message OpOutput { + OutputDataType data_type = 1; + OutputFormat format = 2; + Shape shape = 3; + OriginalOp original_op = 4; // the original op corresponding to the output + bytes data = 5; + uint64 size = 6; +} + +message OpInput { + OutputDataType data_type = 1; + OutputFormat format = 2; + Shape shape = 3; + bytes data = 4; + uint64 size = 5; +} + +enum BufferType { + L1 = 0; +} + +message OpBuffer { + BufferType buffer_type = 1; + bytes data = 2; + uint64 size = 3; +} + +message DumpData{ + string version = 1; + uint64 dump_time = 2; + repeated OpOutput output = 3; + repeated OpInput input = 4; + repeated OpBuffer buffer = 5; +} diff --git a/ge/proto/fusion_model.proto b/ge/proto/fusion_model.proto new file mode 100755 index 00000000..c92c5581 --- /dev/null +++ b/ge/proto/fusion_model.proto @@ -0,0 +1,21 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +import "om.proto"; + +package domi; + +message FusionModelDef { + string version = 1; + repeated OpDef fusion_op = 2; +} \ No newline at end of file diff --git a/ge/proto/fwk_adapter.proto b/ge/proto/fwk_adapter.proto new file mode 100644 index 00000000..9335c926 --- /dev/null +++ b/ge/proto/fwk_adapter.proto @@ -0,0 +1,37 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +package aicpu.FWKAdapter; +option cc_enable_arenas = true; + + +// Defines an struct for input and output. +message TensorDataInfo { + + // value DataType + uint32 dtype = 1; + + // shape dim + repeated int64 dim = 2; + + // data point addr + int64 data_addr = 3; +} + +message KernelRunParam { + // input + repeated TensorDataInfo input = 1; + // output + repeated TensorDataInfo output = 2; +} + diff --git a/ge/proto/ge_api.proto b/ge/proto/ge_api.proto new file mode 100755 index 00000000..331c5aea --- /dev/null +++ b/ge/proto/ge_api.proto @@ -0,0 +1,88 @@ +syntax = "proto3"; +package ge.api_pb; + +import "ge_ir.proto"; + +// GE initialize +message GEInitialize { + map options = 1; +}; + +// initialize response +message GEInitializeResponse { + uint32 status = 1; + uint32 clientId = 2; +}; + +// GE finalize +message GEFinalize { + bool final = 1; + uint32 clientId = 2; +}; + +message GEFinalizeResponse { + uint32 status = 1; +}; + +// GE Session +message CreateSession{ + map options = 1; +}; + +message CreateSessionResponse { + uint32 status = 1; + uint64 sessionId = 2; +}; + +//GE AddGraph +//model serialize :: serializegraph +message SessionAddGraph{ + uint32 graphId = 1; + uint64 sessionId = 2; + ge.proto.GraphDef graph = 3; +}; + +message SessionAddGraphResponse { + uint32 status = 1; +}; + +//GE SessionRemoveGraph +message SessionRemoveGraph{ + uint32 graphId = 1; + uint64 sessionId = 2; +}; + +message SessionRemoveGraphResponse { + uint32 status = 1; +}; + +message SessionRunGraph{ + uint32 graphId = 1; + uint64 sessionId = 2; + repeated ge.proto.TensorDef tensor = 3; +}; + +message SessionBuildGraph{ + uint32 graphId = 1; + uint64 sessionId = 2; + repeated ge.proto.TensorDef tensor = 3; + string savePath = 4; +}; + +message SessionRunGraphResponse { + uint32 status = 1; + repeated ge.proto.TensorDef tensor = 2; +}; + +message SessionBuildGraphResponse { + uint32 status = 1; +}; + +message DestroySession{ + bool final = 1; + uint64 sessionId = 2; +}; + +message DestroySessionResponse { + uint32 status = 1; +}; diff --git a/ge/proto/ge_ir.proto b/ge/proto/ge_ir.proto new file mode 100644 index 00000000..e7bfe0cb --- /dev/null +++ b/ge/proto/ge_ir.proto @@ -0,0 +1,190 @@ +syntax = "proto3"; + +package ge.proto; + +enum DataType +{ + DT_UNDEFINED = 0; // Used to indicate a DataType field has not been set. + DT_FLOAT = 1; // float type + DT_FLOAT16 = 2; // fp16 type + DT_INT8 = 3; // int8 type + DT_UINT8 = 4; // uint8 type + DT_INT16 = 5; // int16 type + DT_UINT16 = 6; // uint16 type + DT_INT32 = 7; // + DT_INT64 = 8; // int64 type + DT_UINT32 = 9; // unsigned int32 + DT_UINT64 = 10; // unsigned int64 + DT_BOOL = 11; // bool type + DT_DOUBLE = 12; // double type + DT_STRING = 13; // string type + DT_DUAL_SUB_INT8 = 14; /**< dual output int8 type */ + DT_DUAL_SUB_UINT8 = 15; /**< dual output uint8 type */ + DT_COMPLEX64 = 16; // complex64 type + DT_COMPLEX128 = 17; // complex128 type + DT_QINT8 = 18; // qint8 type + DT_QINT16 = 19; // qint16 type + DT_QINT32 = 20; // qint32 type + DT_QUINT8 = 21; // quint8 type + DT_QUINT16 = 22; // quint16 type + DT_RESOURCE = 23; // resource type + DT_STRING_REF = 24; // string_ref type + DT_DUAL = 25; /**< dual output type */ +} + +message AttrDef +{ + message ListValue + { + enum ListValueType{ + VT_LIST_NONE = 0; + VT_LIST_STRING = 1; + VT_LIST_INT = 2; + VT_LIST_FLOAT = 3; + VT_LIST_BOOL = 4; + VT_LIST_BYTES = 5; + VT_LIST_TENSOR_DESC = 6; + VT_LIST_TENSOR = 7; + VT_LIST_GRAPH = 8; + VT_LIST_NAMED_ATTRS = 9; + VT_LIST_DATA_TYPE = 10; + } + repeated bytes s = 2; // "list(string)" + repeated int64 i = 3; // "list(int)" + repeated float f = 4; // "list(float)" + repeated bool b = 5; // "list(bool)" + repeated bytes bt = 7; + repeated TensorDescriptor td = 8; + repeated TensorDef t = 9; + repeated GraphDef g = 10; + repeated NamedAttrs na = 11; + repeated int64 dt = 12; // list ge::DataType + + ListValueType val_type = 20; + } + + message ListListInt{ + message ListInt{ + repeated int64 list_i = 1; // list int + } + repeated ListInt list_list_i = 1; // list list int + } + + oneof value + { + bytes s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + bytes bt = 7; + ListValue list = 1; // any "list(...)" + NamedAttrs func = 10; // Used to support attr nesting + TensorDescriptor td = 11; // GeTensorDesc type + TensorDef t = 12; // GeTensor type + GraphDef g = 13; // Graph type + ListListInt list_list_int = 14; // List List Int type + int64 dt = 15; // ge::DataType + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NamedAttrs +{ + string name = 1; + map attr = 2; +} + +// Shape / dimension description, using row-major order +message ShapeDef +{ + repeated int64 dim = 1; // Size of each dimension +} + +// Multidimensional data description +message TensorDescriptor +{ + string name = 1; // Optional parameter, tensor name + + DataType dtype = 2; // tensor datatype + ShapeDef shape = 3; // Shape / dimension + string layout = 4; // Tensor format, eg: "NCHW", "NHWC", "CHW", "ND" + + bool has_out_attr = 9; + int64 size = 10; + int64 weight_size = 11; + bool reuse_input = 12; + bool output_tensor = 13; + string device_type = 14; + bool input_tensor =15; + int64 real_dim_cnt = 16; + int64 reuse_input_index = 17; + int64 data_offset = 18; + int64 cmps_size = 19; + string cmps_tab = 20; + int64 cmps_tab_offset = 21; + + map attr = 5; // Set of extra parameter fields +} + +// GeTensor definition +message TensorDef +{ + TensorDescriptor desc = 1; // Tensor description + bytes data = 2; // Tensor data +} + + +// Operator description +message OpDef +{ + string name = 1; // name + string type = 2; // type + + repeated string input = 5; // input original op name + outgoing index. op_name锛歩ndex + + map attr = 10; // Set of operator parameter fields + + bool has_out_attr = 20; + int64 id = 21; + int64 stream_id =22; + repeated string input_name = 23; + repeated string src_name = 24; + repeated int64 src_index = 25; + repeated string dst_name = 26; + repeated int64 dst_index = 27; + repeated int64 input_i = 28; + repeated int64 output_i = 29; + repeated int64 workspace = 30; + repeated int64 workspace_bytes = 31; + repeated bool is_input_const = 32; + repeated TensorDescriptor input_desc = 33; + repeated TensorDescriptor output_desc = 34; + repeated string subgraph_name = 35; +} + +// Graph definition +message GraphDef +{ + string name = 1; // name + + repeated string input = 4; // Graph input + repeated string output = 5; // Graph output + + repeated OpDef op = 6; // List of operators + + map attr = 11; // Extended field +} + +// model definition +message ModelDef +{ + string name = 1; // name + uint32 version = 2; // IR Proto verion + string custom_version = 3; // User model version number, passed in by user + + repeated GraphDef graph = 7; // Graph definition锛実raph[0] represents the main diagram in modeldef + + map attr = 11; // Extended field +} + diff --git a/ge/proto/insert_op.proto b/ge/proto/insert_op.proto new file mode 100644 index 00000000..c635ca14 --- /dev/null +++ b/ge/proto/insert_op.proto @@ -0,0 +1,136 @@ +syntax = "proto3"; + +package domi; + +message InsertNewOps { + repeated AippOpParams aipp_op = 1; + repeated MultiShapeOpParams multi_shape_op = 2; +} + +message AippOpParams { + enum InputFormat { + UNDEFINED = 0; + YUV420SP_U8 = 1; + XRGB8888_U8 = 2; + RGB888_U8 = 3; + YUV400_U8 = 4; + NC1HWC0DI_FP16 = 5; + NC1HWC0DI_S8 = 6; + ARGB8888_U8 = 7; + YUYV_U8 = 8; + YUV422SP_U8 = 9; + AYUV444_U8 = 10; + RAW10 = 11; + RAW12 = 12; + RAW16 = 13; + RAW24 = 14; + RGB16 = 15; + RGB20 = 16; + RGB24 = 17; + RGB8_IR = 18; + RGB16_IR = 19; + RGB24_IR = 20; + } + + enum AippMode { + undefined = 0; + static = 1; + dynamic = 2; + } + + // AIPP模式，区分静态AIPP和动态AIPP + AippMode aipp_mode = 1; + + // related_input_rank参数为必填，类型为整型，配置范围>=0, <=输入Data算子的个数，默认值为0。 + // 标识对模型的第几个输入做AIPP处理，例如模型有两个输入，需要对第2个输入做AIPP，则配置related_input_rank为1。 + uint32 related_input_rank = 2; + + // input_edge_idx参数为可选，类型为整型，配置范围为>=0。 + // 配置该参数的作用，在于对Data算子不同的输出做不同的AIPP处理，如果该参数没有配置，默认对related_input_rank指定的模型输入的所有输出边做AIPP。 + // 配置值 <= Data算子输出边的个数。 + repeated uint32 input_edge_idx = 3; + + // [Begin] 动态AIPP参数，配置静态AIPP时无效 + uint32 max_src_image_size = 4; + + // 是否支持旋转。默认不支持，开启支持旋转时，会有额外的空间和性能损失 + bool support_rotation = 5; + + // [End] 动态AIPP参数 + + + // [Begin] 静态AIPP参数，配置动态AIPP时无效 + InputFormat input_format = 51; + bool csc_switch = 52; + float cpadding_value = 53; + bool rbuv_swap_switch = 54; + bool ax_swap_switch = 55; + bool single_line_mode = 56; + + int32 src_image_size_w = 57; + int32 src_image_size_h = 58; + + bool crop = 59; + int32 load_start_pos_w = 60; + int32 load_start_pos_h = 61; + int32 crop_size_w = 62; + int32 crop_size_h = 63; + + bool resize = 64; + int32 resize_output_w = 65; + int32 resize_output_h = 66; + + bool padding = 67; + int32 left_padding_size = 68; + int32 right_padding_size = 69; + int32 top_padding_size = 70; + int32 bottom_padding_size = 71; + + int32 mean_chn_0 = 10; + int32 mean_chn_1 = 11; + int32 mean_chn_2 = 12; + int32 mean_chn_3 = 19; + float min_chn_0 = 13; + float min_chn_1 = 14; + float min_chn_2 = 15; + float min_chn_3 = 20; + repeated float var_reci_chn_0 = 16; + repeated float var_reci_chn_1 = 17; + repeated float var_reci_chn_2 = 18; + repeated float var_reci_chn_3 = 21; + + repeated int32 matrix_r0c0 = 30; + repeated int32 matrix_r0c1 = 31; + repeated int32 matrix_r0c2 = 32; + repeated int32 matrix_r1c0 = 33; + repeated int32 matrix_r1c1 = 34; + repeated int32 matrix_r1c2 = 35; + repeated int32 matrix_r2c0 = 36; + repeated int32 matrix_r2c1 = 37; + repeated int32 matrix_r2c2 = 38; + repeated int32 output_bias_0 = 39; + repeated int32 output_bias_1 = 40; + repeated int32 output_bias_2 = 41; + repeated int32 input_bias_0 = 42; + repeated int32 input_bias_1 = 43; + repeated int32 input_bias_2 = 44; + + // [End] 静态AIPP参数 + + // The n number that is used for raw/rgbir data into f16 transformation. + // The transformation equation is x/(2^n). If set to 0, no transform is performed. + uint32 raw_rgbir_to_f16_n = 45; +} + +message MultiShapeOpParams { + enum MultiShapeMode { + batch = 0; //动态batch + resolution = 1; //动态分辨率，扩展用 + } + + MultiShapeMode mode = 1; //算子模式 + uint32 related_input_rank = 2; //新增算子插入到哪个输入 + + + repeated uint32 batch_list = 11; //batch_list值，batch_list的个数是2到8之间 +} diff --git a/ge/proto/om.proto b/ge/proto/om.proto new file mode 100644 index 00000000..e15e5f80 --- /dev/null +++ b/ge/proto/om.proto @@ -0,0 +1,396 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +package domi; + +enum TargetType +{ + MINI = 0; + TINY = 1; + LITE = 2; +} + +// offline model +message ModelDef { + string name = 1; + uint32 version = 2; + + uint64 memory_size = 10; + uint32 stream_num = 11; + uint32 event_num = 12; + uint64 weight_size = 13; + uint32 label_num = 15; + repeated OpDef op = 20; + TargetType target_type = 23; + + map attr = 30; +}; + +// operator define +message OpDef { + string name = 1; + string type = 2; + + uint32 id = 3; + uint32 stream_id = 4; + + repeated string input_name = 5; + + repeated string src_name = 8; + repeated int32 src_index = 9; + repeated int64 input = 10; + repeated int64 output = 11; + repeated TensorDescriptor input_desc = 12; + repeated TensorDescriptor output_desc = 13; + repeated WeightDef weights = 14; + repeated string dst_name = 15; + repeated int32 dst_index = 16; + + repeated int64 workspace = 20; + repeated uint32 workspace_bytes = 21; + + repeated string weight_name = 22; + repeated bool is_input_const = 23; + + map attr = 30; + + QuantizeFactorParams quantize_factor = 31; + + oneof op_params { + // start at 100 here + SendOpParams sender_param = 100; + RecvOpParams receiver_param = 200; + ConvolutionOpParams convolution_param = 300; + PoolingOpParams pooling_param = 400; + EltwiseOpParams eltwise_param = 500; + BatchNormOpParams batchnorm_param = 600; + ScaleOpParams scale_param = 700; + FullConnectionOpParams full_connection_param = 800; + SoftmaxOpParams softmax_param = 900; + ActivationOpParams activation_param = 1000; + ReshapeOpParams reshape_param = 1100; + } +}; + +message SendOpParams { + uint32 event_id = 1; +}; + +message RecvOpParams { + uint32 event_id = 1; +}; + +enum QuantizeScaleType +{ + VECTOR_SCALE = 0; + SCALAR_SCALE = 1; +} + +enum QuantizeScaleMode +{ + NORMAL_MODE = 0; + SQRT_MODE = 1; +} + +enum QuantizeAlgorithm +{ + NON_OFFSET_ALGO = 0; + HALF_OFFSET_ALGO = 1; + ALL_OFFSET_ALGO = 2; +} +message QuantizeFactor +{ + QuantizeScaleMode scale_mode = 1; + bytes scale_value = 2; + int64 scale_offset = 3; + bytes offset_data_value = 4; + int64 offset_data_offset = 5; + bytes offset_weight_value = 6; + int64 offset_weight_offset = 7; + bytes offset_pad_value = 8; + int64 offset_pad_offset = 9; +}; + +message QuantizeCalcFactor +{ + bytes offsetw = 1; + int64 offsetw_offset = 2; + bytes offsetd = 3; + int64 offsetd_offset = 4; + bytes scalereq = 5; + int64 scaledreq_offset = 6; + bytes offsetdnext = 7; + int64 offsetdnext_offset = 8; +} + +message QuantizeFactorParams +{ + QuantizeAlgorithm quantize_algo = 1; + QuantizeScaleType scale_type = 2; + QuantizeFactor quantize_param = 3; + QuantizeFactor dequantize_param = 4; + QuantizeFactor requantize_param = 5; + QuantizeCalcFactor quantizecalc_param = 6; +}; + +message ConvolutionOpParams { + int32 mode = 1; + int32 algo = 2; + int32 pad_mode = 3; + uint32 group = 4; + uint32 num_output = 5; + + repeated uint32 pad = 10; + repeated uint32 stride = 11; + repeated uint32 dilation = 12; + repeated uint32 kernel = 13; + + float alpha = 20; + float beta = 21; + + WeightDef filter = 40; + WeightDef bias = 41; + + bool relu_flag = 62; + repeated uint32 adj = 70; + repeated uint32 target_shape = 71; + repeated uint32 before_pad = 72; +}; + +message PoolingOpParams { + int32 mode = 1; + int32 nan_opt = 2; + int32 pad_mode = 3; + bool global_pooling = 4; + + repeated uint32 window = 10; + repeated uint32 pad = 11; + repeated uint32 stride = 12; + bool ceil_mode = 13; + int32 data_mode = 14; + + float alpha = 20; + float beta = 21; + repeated uint32 before_pad = 22; +}; + +message EltwiseOpParams { + int32 mode = 1; + repeated float coeff = 2; + float alpha = 3; + float beta = 4; + repeated WeightDef weight = 5; + bool relu_flag = 6; +}; + +message ActivationOpParams { + int32 mode = 1; + float coef = 2; + float alpha = 3; + float beta = 4; +}; + +message BatchNormOpParams { + int32 mode = 1; + + float alpha = 2; + float beta = 3; + double epsilon = 4;//optinal,[default = 1e-5] + bool use_global_stats = 5; //optinal,by default true,testing mode + float moving_average_fraction = 6; //optinal,[default = .999]; + + WeightDef estimated_mean = 7; + WeightDef estimated_variance = 8; + + WeightDef scale = 9; + WeightDef bias = 10; +}; + +message ScaleOpParams { + WeightDef scale = 1; + WeightDef bias = 2; +}; + +message ReshapeOpParams { + float alpha = 1; + float beta = 2; + ShapeDef shape = 3; + int32 axis = 4; + int32 num_axes = 5; + int32 format = 6; +}; + +message SoftmaxOpParams { + int32 algo = 1; + int32 mode = 2; + float alpha = 3; + float beta = 4; +}; + +message FullConnectionOpParams { + WeightDef filter = 1; + WeightDef bias = 2; + uint32 num_output = 3; + bool relu_flag = 12; +}; + +message FlattenOpParams { + float alpha = 1; + float beta = 2; + int32 start_axis = 3; + int32 end_axis = 4; +} + +message AddLimitedOpParams { + float alpha = 1; + float beta = 2; + int32 axis = 3; + bool broadcast = 4; + + repeated WeightDef weight = 10; +}; + +message MulLimitedOpParams { + float alpha = 1; + float beta = 2; + int32 axis = 3; + bool broadcast = 4; + + repeated WeightDef weight = 10; +}; + +message AddOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message MulOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message SubOpParams { + float alpha = 1; + float beta = 2; + + repeated WeightDef weight = 10; +}; + +message BiasAddOpParams { + float alpha = 1; + float beta = 2; + + WeightDef bias = 10; +}; + +message MatMulOpParams { + float alpha = 1; + float beta = 2; + bool transposeX = 3; + bool transposeW = 4; + + WeightDef filter = 10; + WeightDef bias = 12; +}; + +message RsqrtOpParams { + float alpha = 1; + float beta = 2; +}; + + +message WeightDef { + int32 format = 1; + int32 data_type = 2; + ShapeDef shape = 3; + bytes data = 4; + int64 data_offset = 5; + uint32 cmps_size = 6; + bytes cmps_tab = 7; + int64 cmps_tab_offset = 10; + CompressInfo cmps_info = 8; + AllOffsetQuantizeInfo alloffset_quantize_info = 11; +} + +message ShapeDef { + repeated int64 dim = 1; +} + +enum DeviceType { + NPU = 0; // In default, we will use NPU. + CPU = 1; // CPU +} + +message AllOffsetQuantizeInfo { + float scale = 1; + int32 offset = 2; +} + +message TensorDescriptor { + int32 format = 1; + int32 data_type = 2; + repeated int64 dim = 3; + uint32 size = 4; + bool reuse_input = 5; + bool output_tensor = 7; + DeviceType device_type = 8; + bool input_tensor = 9; + uint32 real_dim_cnt = 10; + uint32 reuse_input_index = 11; + AllOffsetQuantizeInfo alloffset_quantize_info = 12; +} + +message CompressInfo { + int32 blockRow = 1; // block row + int32 blockCol = 2; // block col + int32 fractalK = 3; // fractal K + int32 fractalN = 4; // fractal N + int32 lastFractalK = 5; // K of last fractal + int32 lastFractalN = 6; // N of last fractal + int32 cubeSize = 7; // cube's length + int32 loadDir = 8; // data load directtiono 0:col load 1:row load +} + +message AttrDef { + message ListValue { + repeated string s = 2; // "list(string)" + repeated int64 i = 3 [packed = true]; // "list(int)" + repeated float f = 4 [packed = true]; // "list(float)" + repeated bool b = 5 [packed = true]; // "list(bool)" + repeated uint32 u = 6 [packed = true]; // "list(uint)" + repeated bytes bt = 7; + } + + oneof value { + string s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + uint32 u = 6; // "uint32" + bytes bt = 7; + ListValue list = 1; // any "list(...)" + NamedAttrs func = 10; + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NamedAttrs { + string name = 1; + map attr = 2; +} + diff --git a/ge/proto/op_mapping_info.proto b/ge/proto/op_mapping_info.proto new file mode 100644 index 00000000..e23b7ebe --- /dev/null +++ b/ge/proto/op_mapping_info.proto @@ -0,0 +1,73 @@ +syntax = "proto3"; +package aicpu.dump; + +message Shape { + repeated uint64 dim = 1; +} + +message Output { + int32 data_type = 1; + int32 format = 2; + Shape shape = 3; + uint64 address = 4; + string original_name = 5; + int32 original_output_index = 6; + int32 original_output_data_type = 7; + int32 original_output_format = 8; + uint64 size = 9; +} + +message Input { + int32 data_type =1; + int32 format = 2; + Shape shape = 3; + uint64 address = 4; + uint64 size = 5; +} + +enum BufferType { + L1 = 0; +} + +message OpBuffer { + BufferType buffer_type = 1; + uint64 address = 2; + uint64 size = 3; +} + +message Op { + string op_name = 1; + string op_type = 2; +} + +message Task { + uint32 task_id = 1; + uint32 stream_id = 2; + Op op = 3; + repeated Output output = 4; + bool end_graph = 5; + repeated Input input = 6; + repeated OpBuffer buffer = 7; +} + +message OpMappingInfo { + string dump_path = 1; + oneof model_name_param { + string model_name = 2; + } + oneof model_id_param { + uint32 model_id = 3; + } + oneof step_id { + uint64 step_id_addr = 4; + } + oneof iterations_per_loop { + uint64 iterations_per_loop_addr = 5; + } + oneof loop_cond { + uint64 loop_cond_addr = 6; + } + uint32 flag = 7; // 0x01 load, 0x00 unload + repeated Task task = 8; + string dump_step = 9; +} \ No newline at end of file diff --git a/ge/proto/optimizer_priority.proto b/ge/proto/optimizer_priority.proto new file mode 100644 index 00000000..769619cf --- /dev/null +++ b/ge/proto/optimizer_priority.proto @@ -0,0 +1,7 @@ +syntax = "proto3"; +package ge.optimizers; + +// Default: GE>FE>AICPU +message Priority{ + repeated string optimizer = 1; +} \ No newline at end of file diff --git a/ge/proto/task.proto b/ge/proto/task.proto new file mode 100644 index 00000000..d0c09840 --- /dev/null +++ b/ge/proto/task.proto @@ -0,0 +1,165 @@ +/* Copyright (C) 2018. Huawei Technologies Co., Ltd. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the Apache License Version 2.0.You may not use this file except in compliance with the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * Apache License for more details at + * http://www.apache.org/licenses/LICENSE-2.0 + */ +syntax = "proto3"; + +package domi; + +message ModelTaskDef { + string version = 1; + + map attr = 9; // Extended field + repeated TaskDef task = 10; + + uint64 memory_size = 11; + uint32 stream_num = 12; + uint32 event_num = 13; + uint64 weight_size = 14; + + repeated bytes op = 15; // input/output opdef in bytes + + uint64 base_addr = 16; // base addr + uint64 weight_addr = 17; // weight addr + uint32 batch_num = 18; +} + + +message TaskDef { + uint32 id = 1; + uint32 type = 2; + + uint32 stream_id = 10; + uint32 event_id = 11; + + KernelDef kernel = 20; + KernelExDef kernel_ex = 21; + KernelHcclDef kernel_hccl = 25; + EventExDef event_ex = 26; + LogTimeStampDef log_timestamp = 28; + + uint32 label_id = 30; + + MemcpyAsyncDef memcpy_async = 31; + StreamSwitchDef stream_switch = 32; + StreamActiveDef stream_active = 33; + bytes private_def = 34; + uint64 ops_kernel_store_ptr = 35; // adjustments to other fields in the future + StreamSwitchNDef stream_switch_n = 36; + + LabelSetDef label_set = 37; + LabelGotoExDef label_goto_ex = 38; + LabelSwitchByIndexDef label_switch_by_index = 39; +} + +message KernelDef { + KernelContext context = 1; + + string stub_func = 10; + uint32 block_dim = 11; + uint32 args_size = 12; + bytes args = 13; + bytes sm_desc = 14; + bytes flowtable = 15; + string so_name = 16; + string kernel_name = 17; + bytes kernel_ext_info = 18; + uint32 kernel_ext_info_size = 19; +} + +message KernelContext { + uint32 kernel_type = 1; + uint32 op_id = 2; // OP type in CCE + uint32 kernel_func_id = 3; + uint32 op_index = 4; // TE/Custom operator + bool is_flowtable = 5; // Identify whether args is a flowtable structure + bytes args_offset = 6; // args offset information + uint32 args_count = 7; // args count + repeated uint32 origin_op_index = 8; +} + + +message KernelExDef { + uint32 flags = 1; + + uint32 op_index = 4; + uint32 args_size = 12; + bytes args = 13; + bytes task_info = 14; // serialized nodeDef, funcDef, inputoutput + uint32 task_info_size = 15; + bytes kernel_ext_info = 16; + uint32 kernel_ext_info_size = 17; +} + + +message KernelHcclDef { + uint32 op_index = 8; + string hccl_type = 9; +} + + +message EventExDef { + uint32 op_index = 1; + uint32 event_type = 2; +} + +message LogTimeStampDef { + uint64 logid = 1; + bool notify = 2; + uint32 flat = 3; +} + +message MemcpyAsyncDef { + uint64 dst = 1; + uint64 dst_max = 2; + uint64 src = 3; + uint64 count = 4; + uint32 kind = 5; + uint32 op_index = 6; +} + +message StreamSwitchDef { + uint32 op_index = 1; + uint32 true_stream_id = 2; + int64 value = 3; + uint64 value_ptr = 4; + uint32 data_type = 5; +} + +message StreamActiveDef { + uint32 op_index = 1; + uint32 active_stream_id = 2; +} + +message StreamSwitchNDef { + uint32 op_index = 1; + uint32 size = 2; + repeated int64 target_value = 3; + repeated uint32 true_stream_id = 4; + uint32 element_size = 5; + uint32 data_type = 6; +} + +message LabelSetDef { + uint32 op_index = 1; + uint32 label_id = 2; + uint32 model_id = 3; +} + +message LabelGotoExDef { + uint32 op_index = 1; + uint32 label_id = 2; + uint32 model_id = 3; +} + +message LabelSwitchByIndexDef { + uint32 op_index = 1; + uint32 label_max = 2; +} diff --git a/ge/proto/tensorflow/attr_value.proto b/ge/proto/tensorflow/attr_value.proto new file mode 100644 index 00000000..1cc67d62 --- /dev/null +++ b/ge/proto/tensorflow/attr_value.proto @@ -0,0 +1,62 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "AttrValueProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "tensor.proto"; +import "tensor_shape.proto"; +import "types.proto"; + +// Protocol buffer representing the value for an attr used to configure an Op. +// Comment indicates the corresponding attr type. Only the field matching the +// attr type may be filled. +message AttrValue { + // LINT.IfChange + message ListValue { + repeated bytes s = 2; // "list(string)" + repeated int64 i = 3 [packed = true]; // "list(int)" + repeated float f = 4 [packed = true]; // "list(float)" + repeated bool b = 5 [packed = true]; // "list(bool)" + repeated DataType type = 6 [packed = true]; // "list(type)" + repeated TensorShapeProto shape = 7; // "list(shape)" + repeated TensorProto tensor = 8; // "list(tensor)" + repeated NameAttrList func = 9; // "list(attr)" + } + // LINT.ThenChange(https://www.tensorflow.org/code/tensorflow/c/c_api.cc) + + oneof value { + bytes s = 2; // "string" + int64 i = 3; // "int" + float f = 4; // "float" + bool b = 5; // "bool" + DataType type = 6; // "type" + TensorShapeProto shape = 7; // "shape" + TensorProto tensor = 8; // "tensor" + ListValue list = 1; // any "list(...)" + + // "func" represents a function. func.name is a function's name or + // a primitive op's name. func.attr.first is the name of an attr + // defined for that function. func.attr.second is the value for + // that attr in the instantiation. + NameAttrList func = 10; + + // This is a placeholder only used in nodes defined inside a + // function. It indicates the attr value will be supplied when + // the function is instantiated. For example, let us suppose a + // node "N" in function "FN". "N" has an attr "A" with value + // placeholder = "foo". When FN is instantiated with attr "foo" + // set to "bar", the instantiated node N's attr A will have been + // given the value "bar". + string placeholder = 9; + } +} + +// A list of attr names and their values. The whole list is attached +// with a string name. E.g., MatMul[T=float]. +message NameAttrList { + string name = 1; + map attr = 2; +} diff --git a/ge/proto/tensorflow/function.proto b/ge/proto/tensorflow/function.proto new file mode 100644 index 00000000..075897c6 --- /dev/null +++ b/ge/proto/tensorflow/function.proto @@ -0,0 +1,100 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "FunctionProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "attr_value.proto"; +import "node_def.proto"; +import "op_def.proto"; + +// A library is a set of named functions. +message FunctionDefLibrary { + repeated FunctionDef function = 1; + repeated GradientDef gradient = 2; +} + +// A function can be instantiated when the runtime can bind every attr +// with a value. When a GraphDef has a call to a function, it must +// have binding for every attr defined in the signature. +// * device spec, etc. +message FunctionDef { + // The definition of the function's name, arguments, return values, + // attrs etc. + OpDef signature = 1; + + // Attributes specific to this function definition. + map attr = 5; + + // NOTE: field id 2 deleted on Jan 11, 2017, GraphDef version 21. + reserved 2; + + // In both of the following fields, there is the need to specify an + // output that is used as either the input to another node (in + // `node_def`) or as a return value of the function (in `ret`). + // Unlike the NodeDefs in GraphDef, we need to be able to specify a + // list in some cases (instead of just single outputs). Also, we + // need to be able to deal with lists of unknown length (so the + // output index may not be known at function definition time). So + // we use the following format instead: + // * "fun_in" where "fun_in" is the name of a function input arg in + // the `signature` field above. This represents that input, whether + // it is a single tensor or a list. + // * "fun_in:0" gives the first element of a function input arg (a + // non-list input is considered a list of length 1 for these + // purposes). + // * "node:out" where "node" is the name of a node in `node_def` and + // "out" is the name one of its op's output arguments (the name + // comes from the OpDef of the node's op). This represents that + // node's output, whether it is a single tensor or a list. + // Note: We enforce that an op's output arguments are never + // renamed in the backwards-compatibility test. + // * "node:out:0" gives the first element of a node output arg (a + // non-list output is considered a list of length 1 for these + // purposes). + // + // NOT CURRENTLY SUPPORTED (but may be in the future): + // * "node:out:-1" gives last element in a node output list + // * "node:out:1:" gives a list with all but the first element in a + // node output list + // * "node:out::-1" gives a list with all but the last element in a + // node output list + + // The body of the function. Unlike the NodeDefs in a GraphDef, attrs + // may have values of type `placeholder` and the `input` field uses + // the "output" format above. + + // By convention, "op" in node_def is resolved by consulting with a + // user-defined library first. If not resolved, "func" is assumed to + // be a builtin op. + repeated NodeDef node_def = 3; + + // A mapping from the output arg names from `signature` to the + // outputs from `node_def` that should be returned by the function. + map ret = 4; +} + +// GradientDef defines the gradient function of a function defined in +// a function library. +// +// A gradient function g (specified by gradient_func) for a function f +// (specified by function_name) must follow the following: +// +// The function 'f' must be a numerical function which takes N inputs +// and produces M outputs. Its gradient function 'g', which is a +// function taking N + M inputs and produces N outputs. +// +// I.e. if we have +// (y1, y2, ..., y_M) = f(x1, x2, ..., x_N), +// then, g is +// (dL/dx1, dL/dx2, ..., dL/dx_N) = g(x1, x2, ..., x_N, +// dL/dy1, dL/dy2, ..., dL/dy_M), +// where L is a scalar-value function of (x1, x2, ..., xN) (e.g., the +// loss function). dL/dx_i is the partial derivative of L with respect +// to x_i. +message GradientDef { + string function_name = 1; // The function name. + string gradient_func = 2; // The gradient function's name. +} diff --git a/ge/proto/tensorflow/graph.proto b/ge/proto/tensorflow/graph.proto new file mode 100644 index 00000000..d639a7d6 --- /dev/null +++ b/ge/proto/tensorflow/graph.proto @@ -0,0 +1,56 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "GraphProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "node_def.proto"; +import "function.proto"; +import "versions.proto"; + +// Represents the graph of operations +message GraphDef { + repeated NodeDef node = 1; + + // Compatibility versions of the graph. See core/public/version.h for version + // history. The GraphDef version is distinct from the TensorFlow version, and + // each release of TensorFlow will support a range of GraphDef versions. + VersionDef versions = 4; + + // Deprecated single version field; use versions above instead. Since all + // GraphDef changes before "versions" was introduced were forward + // compatible, this field is entirely ignored. + int32 version = 3 [deprecated = true]; + + // EXPERIMENTAL. DO NOT USE OR DEPEND ON THIS YET. + // + // "library" provides user-defined functions. + // + // Naming: + // * library.function.name are in a flat namespace. + // NOTE: We may need to change it to be hierarchical to support + // different orgs. E.g., + // { "/google/nn", { ... }}, + // { "/google/vision", { ... }} + // { "/org_foo/module_bar", { ... }} + // map named_lib; + // * If node[i].op is the name of one function in "library", + // node[i] is deemed as a function call. Otherwise, node[i].op + // must be a primitive operation supported by the runtime. + // + // + // Function call semantics: + // + // * The callee may start execution as soon as some of its inputs + // are ready. The caller may want to use Tuple() mechanism to + // ensure all inputs are ready in the same time. + // + // * The consumer of return values may start executing as soon as + // the return values the consumer depends on are ready. The + // consumer may want to use Tuple() mechanism to ensure the + // consumer does not start until all return values of the callee + // function are ready. + FunctionDefLibrary library = 2; +}; diff --git a/ge/proto/tensorflow/graph_library.proto b/ge/proto/tensorflow/graph_library.proto new file mode 100644 index 00000000..e393d38d --- /dev/null +++ b/ge/proto/tensorflow/graph_library.proto @@ -0,0 +1,14 @@ +syntax = "proto3"; + +package domi.tensorflow; + +import "graph.proto"; + +message GeGraphDef { + string name = 1; + GraphDef graph = 2; +} + +message GraphDefLibrary { + repeated GeGraphDef graph_def = 1; +}; \ No newline at end of file diff --git a/ge/proto/tensorflow/node_def.proto b/ge/proto/tensorflow/node_def.proto new file mode 100644 index 00000000..b9bc97ee --- /dev/null +++ b/ge/proto/tensorflow/node_def.proto @@ -0,0 +1,63 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "NodeProto"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "attr_value.proto"; + +message NodeDef { + // The name given to this operator. Used for naming inputs, + // logging, visualization, etc. Unique within a single GraphDef. + // Must match the regexp "[A-Za-z0-9.][A-Za-z0-9_./]*". + string name = 1; + + // The operation name. There may be custom parameters in attrs. + // Op names starting with an underscore are reserved for internal use. + string op = 2; + + // Each input is "node:src_output" with "node" being a string name and + // "src_output" indicating which output tensor to use from "node". If + // "src_output" is 0 the ":0" suffix can be omitted. Regular inputs + // may optionally be followed by control inputs that have the format + // "^node". + repeated string input = 3; + + // A (possibly partial) specification for the device on which this + // node should be placed. + // The expected syntax for this string is as follows: + // + // DEVICE_SPEC ::= PARTIAL_SPEC + // + // PARTIAL_SPEC ::= ("/" CONSTRAINT) * + // CONSTRAINT ::= ("job:" JOB_NAME) + // | ("replica:" [1-9][0-9]*) + // | ("task:" [1-9][0-9]*) + // | ("device:" [A-Za-z]* ":" ([1-9][0-9]* | "*") ) + // + // Valid values for this string include: + // * "/job:worker/replica:0/task:1/device:GPU:3" (full specification) + // * "/job:worker/device:GPU:3" (partial specification) + // * "" (no specification) + // + // If the constraints do not resolve to a single device (or if this + // field is empty or not present), the runtime will attempt to + // choose a device automatically. + string device = 4; + + // Operation-specific graph-construction-time configuration. + // Note that this should include all attrs defined in the + // corresponding OpDef, including those with a value matching + // the default -- this allows the default to change and makes + // NodeDefs easier to interpret on their own. However, if + // an attr with a default is not specified in this list, the + // default will be used. + // The "names" (keys) must match the regexp "[a-z][a-z0-9_]+" (and + // one of the names from the corresponding OpDef's attr field). + // The values must have a type matching the corresponding OpDef + // attr's type field. + // Add some examples here showing best practices. + map attr = 5; +}; diff --git a/ge/proto/tensorflow/op_def.proto b/ge/proto/tensorflow/op_def.proto new file mode 100644 index 00000000..3485d045 --- /dev/null +++ b/ge/proto/tensorflow/op_def.proto @@ -0,0 +1,164 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "OpDefProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "attr_value.proto"; +import "types.proto"; + +// Defines an operation. A NodeDef in a GraphDef specifies an Op by +// using the "op" field which should match the name of a OpDef. +// LINT.IfChange +message OpDef { + // Op names starting with an underscore are reserved for internal use. + // Names should be CamelCase and match the regexp "[A-Z][a-zA-Z0-9_]*". + string name = 1; + + // For describing inputs and outputs. + message ArgDef { + // Name for the input/output. Should match the regexp "[a-z][a-z0-9_]*". + string name = 1; + + // Human readable description. + string description = 2; + + // Describes the type of one or more tensors that are accepted/produced + // by this input/output arg. The only legal combinations are: + // * For a single tensor: either the "type" field is set or the + // "type_attr" field is set to the name of an attr with type "type". + // * For a sequence of tensors with the same type: the "number_attr" + // field will be set to the name of an attr with type "int", and + // either the "type" or "type_attr" field will be set as for + // single tensors. + // * For a sequence of tensors, the "type_list_attr" field will be set + // to the name of an attr with type "list(type)". + DataType type = 3; + string type_attr = 4; // if specified, attr must have type "type" + string number_attr = 5; // if specified, attr must have type "int" + // If specified, attr must have type "list(type)", and none of + // type, type_attr, and number_attr may be specified. + string type_list_attr = 6; + + // For inputs: if true, the inputs are required to be refs. + // By default, inputs can be either refs or non-refs. + // For outputs: if true, outputs are refs, otherwise they are not. + bool is_ref = 16; + }; + + // Description of the input(s). + repeated ArgDef input_arg = 2; + + // Description of the output(s). + repeated ArgDef output_arg = 3; + + // Description of the graph-construction-time configuration of this + // Op. That is to say, this describes the attr fields that will + // be specified in the NodeDef. + message AttrDef { + // A descriptive name for the argument. May be used, e.g. by the + // Python client, as a keyword argument name, and so should match + // the regexp "[a-z][a-z0-9_]+". + string name = 1; + + // One of the type names from attr_value.proto ("string", "list(string)", + // "int", etc.). + string type = 2; + + // A reasonable default for this attribute if the user does not supply + // a value. If not specified, the user must supply a value. + AttrValue default_value = 3; + + // Human-readable description. + string description = 4; + + + // --- Constraints --- + // These constraints are only in effect if specified. Default is no + // constraints. + + // For type == "int", this is a minimum value. For "list(___)" + // types, this is the minimum length. + bool has_minimum = 5; + int64 minimum = 6; + + // The set of allowed values. Has type that is the "list" version + // of the "type" field above (uses the "list" field of AttrValue). + // If type == "type" or "list(type)" above, then the "type" field + // of "allowed_values.list" has the set of allowed DataTypes. + // If type == "string" or "list(string)", then the "s" field of + // "allowed_values.list" has the set of allowed strings. + AttrValue allowed_values = 7; + } + repeated AttrDef attr = 4; + + // Optional deprecation based on GraphDef versions. + OpDeprecation deprecation = 8; + + // One-line human-readable description of what the Op does. + string summary = 5; + + // Additional, longer human-readable description of what the Op does. + string description = 6; + + // ------------------------------------------------------------------------- + // Which optimizations this operation can participate in. + + // True if the operation is commutative ("op(a,b) == op(b,a)" for all inputs) + bool is_commutative = 18; + + // If is_aggregate is true, then this operation accepts N >= 2 + // inputs and produces 1 output all of the same type. Should be + // associative and commutative, and produce output with the same + // shape as the input. The optimizer may replace an aggregate op + // taking input from multiple devices with a tree of aggregate ops + // that aggregate locally within each device (and possibly within + // groups of nearby devices) before communicating. + bool is_aggregate = 16; // for things like add + + // Other optimizations go here, like + // can_alias_input, rewrite_when_output_unused, partitioning_strategy, etc. + + // ------------------------------------------------------------------------- + // Optimization constraints. + + // Ops are marked as stateful if their behavior depends on some state beyond + // their input tensors (e.g. variable reading op) or if they have + // a side-effect (e.g. printing or asserting ops). Equivalently, stateless ops + // must always produce the same output for the same input and have + // no side-effects. + // + // By default Ops may be moved between devices. Stateful ops should + // either not be moved, or should only be moved if that state can also + // be moved (e.g. via some sort of save / restore). + // Stateful ops are guaranteed to never be optimized away by Common + // Subexpression Elimination (CSE). + bool is_stateful = 17; // for things like variables, queue + + // ------------------------------------------------------------------------- + // Non-standard options. + + // By default, all inputs to an Op must be initialized Tensors. Ops + // that may initialize tensors for the first time should set this + // field to true, to allow the Op to take an uninitialized Tensor as + // input. + bool allows_uninitialized_input = 19; // for Assign, etc. +}; +// LINT.ThenChange( +// https://www.tensorflow.org/code/tensorflow/core/framework/op_def_util.cc) + +// Information about version-dependent deprecation of an op +message OpDeprecation { + // First GraphDef version at which the op is disallowed. + int32 version = 1; + + // Explanation of why it was deprecated and what to use instead. + string explanation = 2; +}; + +// A collection of OpDefs +message OpList { + repeated OpDef op = 1; +}; diff --git a/ge/proto/tensorflow/resource_handle.proto b/ge/proto/tensorflow/resource_handle.proto new file mode 100644 index 00000000..a3452351 --- /dev/null +++ b/ge/proto/tensorflow/resource_handle.proto @@ -0,0 +1,29 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "ResourceHandle"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +// Protocol buffer representing a handle to a tensorflow resource. Handles are +// not valid across executions, but can be serialized back and forth from within +// a single run. +message ResourceHandleProto { + // Unique name for the device containing the resource. + string device = 1; + + // Container in which this resource is placed. + string container = 2; + + // Unique name of this resource. + string name = 3; + + // Hash code for the type of the resource. Is only valid in the same device + // and in the same execution. + uint64 hash_code = 4; + + // For debug-only, the name of the type pointed to by this handle, if + // available. + string maybe_type_name = 5; +}; diff --git a/ge/proto/tensorflow/tensor.proto b/ge/proto/tensorflow/tensor.proto new file mode 100644 index 00000000..d0a4d024 --- /dev/null +++ b/ge/proto/tensorflow/tensor.proto @@ -0,0 +1,94 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "TensorProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +import "resource_handle.proto"; +import "tensor_shape.proto"; +import "types.proto"; + +// Protocol buffer representing a tensor. +message TensorProto { + DataType dtype = 1; + + // Shape of the tensor. + TensorShapeProto tensor_shape = 2; + + // Only one of the representations below is set, one of "tensor_contents" and + // the "xxx_val" attributes. We are not using oneof because as oneofs cannot + // contain repeated fields it would require another extra set of messages. + + // Version number. + // + // In version 0, if the "repeated xxx" representations contain only one + // element, that element is repeated to fill the shape. This makes it easy + // to represent a constant Tensor with a single value. + int32 version_number = 3; + + // Serialized raw tensor content from either Tensor::AsProtoTensorContent or + // memcpy in tensorflow::grpc::EncodeTensorToByteBuffer. This representation + // can be used for all tensor types. The purpose of this representation is to + // reduce serialization overhead during RPC call by avoiding serialization of + // many repeated small items. + bytes tensor_content = 4; + + // Type specific representations that make it easy to create tensor protos in + // all languages. Only the representation corresponding to "dtype" can + // be set. The values hold the flattened representation of the tensor in + // row major order. + + // DT_HALF, DT_BFLOAT16. Note that since protobuf has no int16 type, we'll + // have some pointless zero padding for each value here. + repeated int32 half_val = 13 [packed = true]; + + // DT_FLOAT. + repeated float float_val = 5 [packed = true]; + + // DT_DOUBLE. + repeated double double_val = 6 [packed = true]; + + // DT_INT32, DT_INT16, DT_INT8, DT_UINT8. + repeated int32 int_val = 7 [packed = true]; + + // DT_STRING + repeated bytes string_val = 8; + + // DT_COMPLEX64. scomplex_val(2*i) and scomplex_val(2*i+1) are real + // and imaginary parts of i-th single precision complex. + repeated float scomplex_val = 9 [packed = true]; + + // DT_INT64 + repeated int64 int64_val = 10 [packed = true]; + + // DT_BOOL + repeated bool bool_val = 11 [packed = true]; + + // DT_COMPLEX128. dcomplex_val(2*i) and dcomplex_val(2*i+1) are real + // and imaginary parts of i-th double precision complex. + repeated double dcomplex_val = 12 [packed = true]; + + // DT_RESOURCE + repeated ResourceHandleProto resource_handle_val = 14; + + // DT_VARIANT + repeated VariantTensorDataProto variant_val = 15; + + // DT_UINT32 + repeated uint32 uint32_val = 16 [packed = true]; + + // DT_UINT64 + repeated uint64 uint64_val = 17 [packed = true]; +}; + +// Protocol buffer representing the serialization format of DT_VARIANT tensors. +message VariantTensorDataProto { + // Name of the type of objects being serialized. + string type_name = 1; + // Portions of the object that are not Tensors. + bytes metadata = 2; + // Tensors contained within objects being serialized. + repeated TensorProto tensors = 3; +} diff --git a/ge/proto/tensorflow/tensor_shape.proto b/ge/proto/tensorflow/tensor_shape.proto new file mode 100644 index 00000000..4225a2e3 --- /dev/null +++ b/ge/proto/tensorflow/tensor_shape.proto @@ -0,0 +1,45 @@ +// Protocol buffer representing the shape of tensors. + +syntax = "proto3"; +option cc_enable_arenas = true; +option java_outer_classname = "TensorShapeProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +package domi.tensorflow; + +// Dimensions of a tensor. +message TensorShapeProto { + // One dimension of the tensor. + message Dim { + // Size of the tensor in that dimension. + // This value must be >= -1, but values of -1 are reserved for "unknown" + // shapes (values of -1 mean "unknown" dimension). Certain wrappers + // that work with TensorShapeProto may fail at runtime when deserializing + // a TensorShapeProto containing a dim value of -1. + int64 size = 1; + + // Optional name of the tensor dimension. + string name = 2; + }; + + // Dimensions of the tensor, such as {"input", 30}, {"output", 40} + // for a 30 x 40 2D tensor. If an entry has size -1, this + // corresponds to a dimension of unknown size. The names are + // optional. + // + // The order of entries in "dim" matters: It indicates the layout of the + // values in the tensor in-memory representation. + // + // The first entry in "dim" is the outermost dimension used to layout the + // values, the last entry is the innermost dimension. This matches the + // in-memory layout of RowMajor Eigen tensors. + // + // If "dim.size()" > 0, "unknown_rank" must be false. + repeated Dim dim = 2; + + // If true, the number of dimensions in the shape is unknown. + // + // If true, "dim.size()" must be 0. + bool unknown_rank = 3; +}; diff --git a/ge/proto/tensorflow/types.proto b/ge/proto/tensorflow/types.proto new file mode 100644 index 00000000..ba7a72b3 --- /dev/null +++ b/ge/proto/tensorflow/types.proto @@ -0,0 +1,74 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "TypesProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +// LINT.IfChange +enum DataType { + // Not a legal value for DataType. Used to indicate a DataType field + // has not been set. + DT_INVALID = 0; + + // Data types that all computation devices are expected to be + // capable to support. + DT_FLOAT = 1; + DT_DOUBLE = 2; + DT_INT32 = 3; + DT_UINT8 = 4; + DT_INT16 = 5; + DT_INT8 = 6; + DT_STRING = 7; + DT_COMPLEX64 = 8; // Single-precision complex + DT_INT64 = 9; + DT_BOOL = 10; + DT_QINT8 = 11; // Quantized int8 + DT_QUINT8 = 12; // Quantized uint8 + DT_QINT32 = 13; // Quantized int32 + DT_BFLOAT16 = 14; // Float32 truncated to 16 bits. Only for cast ops. + DT_QINT16 = 15; // Quantized int16 + DT_QUINT16 = 16; // Quantized uint16 + DT_UINT16 = 17; + DT_COMPLEX128 = 18; // Double-precision complex + DT_HALF = 19; + DT_RESOURCE = 20; + DT_VARIANT = 21; // Arbitrary C++ data types + DT_UINT32 = 22; + DT_UINT64 = 23; + + // Do not use! These are only for parameters. Every enum above + // should have a corresponding value below (verified by types_test). + DT_FLOAT_REF = 101; + DT_DOUBLE_REF = 102; + DT_INT32_REF = 103; + DT_UINT8_REF = 104; + DT_INT16_REF = 105; + DT_INT8_REF = 106; + DT_STRING_REF = 107; + DT_COMPLEX64_REF = 108; + DT_INT64_REF = 109; + DT_BOOL_REF = 110; + DT_QINT8_REF = 111; + DT_QUINT8_REF = 112; + DT_QINT32_REF = 113; + DT_BFLOAT16_REF = 114; + DT_QINT16_REF = 115; + DT_QUINT16_REF = 116; + DT_UINT16_REF = 117; + DT_COMPLEX128_REF = 118; + DT_HALF_REF = 119; + DT_RESOURCE_REF = 120; + DT_VARIANT_REF = 121; + DT_UINT32_REF = 122; + DT_UINT64_REF = 123; +} +// LINT.ThenChange( +// https://www.tensorflow.org/code/tensorflow/c/c_api.h, +// https://www.tensorflow.org/code/tensorflow/go/tensor.go, +// https://www.tensorflow.org/code/tensorflow/core/framework/tensor.cc, +// https://www.tensorflow.org/code/tensorflow/core/framework/types.h, +// https://www.tensorflow.org/code/tensorflow/core/framework/types.cc, +// https://www.tensorflow.org/code/tensorflow/python/framework/dtypes.py, +// https://www.tensorflow.org/code/tensorflow/python/framework/function.py) diff --git a/ge/proto/tensorflow/versions.proto b/ge/proto/tensorflow/versions.proto new file mode 100644 index 00000000..48061218 --- /dev/null +++ b/ge/proto/tensorflow/versions.proto @@ -0,0 +1,31 @@ +syntax = "proto3"; + +package domi.tensorflow; +option cc_enable_arenas = true; +option java_outer_classname = "VersionsProtos"; +option java_multiple_files = true; +option java_package = "org.tensorflow.framework"; + +// Version information for a piece of serialized data +// +// There are different types of versions for each type of data +// (GraphDef, etc.), but they all have the same common shape +// described here. +// +// Each consumer has "consumer" and "min_producer" versions (specified +// elsewhere). A consumer is allowed to consume this data if +// +// producer >= min_producer +// consumer >= min_consumer +// consumer not in bad_consumers +// +message VersionDef { + // The version of the code that produced this data. + int32 producer = 1; + + // Any consumer below this version is not allowed to consume this data. + int32 min_consumer = 2; + + // Specific consumer versions which are disallowed (e.g. due to bugs). + repeated int32 bad_consumers = 3; +};