Skip to content

X
X2Paddle

PaddlePaddle / X2Paddle
大约 1 年 前同步成功

通知 328
Star 698
Fork 167
X
X2Paddle
未验证 提交 ce6ffee2 编写于 作者: S SunAhong1993 提交者： GitHub
浏览文件
操作

Merge pull request #7 from PaddlePaddle/develop 

me
上级 bb7cd6b9 09d35587
隐藏空白更改
内联 并排
Showing with 10090 addition and 13500 deletion
README.md
浏览文件 @ ce6ffee2
......@@ -15,7 +15,7 @@ paddlepaddle >= 1.8.0
**按需安装以下依赖**
tensorflow ： tensorflow == 1.14.0
caffe ： 无
onnx ： onnx == 1.6.0
onnx ： onnx >= 1.6.0
## 安装
### 安装方式一（推荐）
......@@ -58,7 +58,7 @@ x2paddle --framework=paddle2onnx --model=paddle_infer_model_dir --save_dir=onnx_
|--save_dir | 指定转换后的模型保存目录路径 |
|--model | 当framework为tensorflow/onnx时，该参数指定tensorflow的pb模型文件或onnx模型路径 |
|--caffe_proto | **[可选]** 由caffe.proto编译成caffe_pb2.py文件的存放路径，当存在自定义Layer时使用，默认为None |
|--without_data_format_optimization | **[可选]** For TensorFlow, 当指定该参数时，关闭NHWC->NCHW的优化，见[文档Q2](FAQ.md) |
|--without_data_format_optimization | **[可选]** For TensorFlow, 当指定该参数为False时，打开NHWC->NCHW的优化，见[文档Q2](FAQ.md)，默认为True|
|--define_input_shape | **[可选]** For TensorFlow, 当指定该参数时，强制用户输入每个Placeholder的shape，见[文档Q2](FAQ.md) |
|--params_merge | **[可选]** 当指定该参数时，转换完成后，inference_model中的所有模型参数将合并保存为一个文件__params__ |
|--onnx_opset | **[可选]** 当framework为paddle2onnx时，该参数可设置转换为ONNX的OpSet版本，目前支持9、10、11，默认为10 |
......
op_list.md
浏览文件 @ ce6ffee2
# X2Paddle支持OP列表
> 目前X2Paddle支持50+的TensorFlow OP，30+的Caffe Layer，覆盖了大部分CV分类模型常用的操作。我们在如下列表中给出了目前X2Paddle支持的全部OP。
> 目前X2Paddle支持70+的TensorFlow OP，30+的Caffe Layer，覆盖了大部分CV分类模型常用的操作。我们在如下列表中给出了目前X2Paddle支持的全部OP。
**注：** 目前，部分OP暂未支持，如您在转换过程中出现OP不支持的情况，可自行添加或反馈给我们。欢迎通过[ISSUE反馈](https://github.com/PaddlePaddle/X2Paddle/issues/new)的方式告知我们(模型名，代码实现或模型获取方式)，我们会及时跟进：）
......@@ -7,20 +7,24 @@
| 序号 | OP | 序号 | OP | 序号 | OP | 序号 | OP |
|------|------|------|------|------|------|------|------|
| 1 | Relu | 2 | Relu6 | 3 | Shape | 4 | Abs |
| 5 | Sigmoid | 6 | Exp | 7 | Rsqrt | 8 | swish_f32 |
| 9 | Tanh | 10 | LeakyRelu | 11 | Add | 12 | RealDiv |
| 13 | Sub | 14 | Maximum | 15 | Mul | 16 | FloorDiv |
| 17 | Placeholder | 18 | Const | 19 | Transpose | 20 | FusedBatchNorm |
| 21 | Conv2D | 22 | BiasAdd | 23 | MaxPool | 24 | DepthwiseConv2dNative |
| 25 | Reshape | 26 | AvgPool | 27 | SplitV | 28 | SquaredDifference |
| 29 | Tile | 30 | Pack | 31 | Pad | 32 | ResizeBilinear |
| 33 | Mean | 34 | MatMul | 35 | ArgMax | 36 | StridedSlice |
| 37 | Slice | 38 | Sum | 39 | Max | 40 | Conv2DBackpropInput |
| 41 | Cast | 42 | Split | 43 | Squeeze | 44 | ResizeNearestNeighbor |
| 45 | Softmax | 46 | Range | 47 | ConcatV2 | 48 | MirrorPad |
| 49 | Identity | 50 | GreaterEqual | 51 | StopGradient | 52 | Minimum |
| 53 | RadnomUniform | 54 | Fill | 55 | Floor | 56 | DepthToSpace |
| 1 | Relu | 2 | Relu6 | 3 | Shape | 4 | Abs |
| 5 | Sigmoid | 6 | Exp | 7 | Rsqrt | 8 | swish_f32 |
| 9 | Tanh | 10 | LeakyRelu | 11 | Add | 12 | RealDiv |
| 13 | Sub | 14 | Maximum | 15 | Mul | 16 | FloorDiv |
| 17 | Placeholder | 18 | Const | 19 | Transpose | 20 | FusedBatchNorm |
| 21 | Conv2D | 22 | BiasAdd | 23 | MaxPool | 24 | DepthwiseConv2dNative |
| 25 | Reshape | 26 | AvgPool | 27 | SplitV | 28 | SquaredDifference |
| 29 | Tile | 30 | Pack | 31 | Pad | 32 | ResizeBilinear |
| 33 | Mean | 34 | MatMul | 35 | ArgMax | 36 | StridedSlice |
| 37 | Slice | 38 | Sum | 39 | Max | 40 | Conv2DBackpropInput |
| 41 | Cast | 42 | Split | 43 | Squeeze | 44 | ResizeNearestNeighbor |
| 45 | Softmax | 46 | Range | 47 | ConcatV2 | 48 | MirrorPad |
| 49 | Identity | 50 | GreaterEqual | 51 | StopGradient | 52 | Minimum |
| 53 | RadnomUniform | 54 | Fill | 55 | Floor | 56 | DepthToSpace |
| 57 | Sqrt | 58 | Softplus | 59 | Erf | 60 | AddV2 |
| 61 | LessEqual | 62 | BatchMatMul | 63 | BatchMatMulV2 | 64 | ExpandDims |
| 65 | BatchToSpaceND | 66 | SpaceToBatchND | 67 | OneHot | 68 | Pow |
| 69 | All | 70 | GatherV2 | 71 | IteratorV2 | | |
## Caffe
......
x2paddle/__init__.py
浏览文件 @ ce6ffee2
__version__ = "0.8.1"
__version__ = "0.8.4"
from .core.program import PaddleProgram
......
x2paddle/convert.py
浏览文件 @ ce6ffee2
......@@ -66,8 +66,8 @@ def arg_parser():
parser.add_argument(
"--without_data_format_optimization",
"-wo",
action="store_true",
default=False,
type=_text_type,
default="True",
help="tf model conversion without data format optimization")
parser.add_argument(
"--define_input_shape",
......@@ -93,7 +93,7 @@ def arg_parser():
def tf2paddle(model_path,
save_dir,
without_data_format_optimization=False,
without_data_format_optimization,
define_input_shape=False,
params_merge=False):
# check tensorflow installation and version
......@@ -170,8 +170,8 @@ def onnx2paddle(model_path, save_dir, params_merge=False):
try:
import onnx
version = onnx.version.version
if version != '1.6.0':
print("[ERROR] onnx==1.6.0 is required")
if version < '1.6.0':
print("[ERROR] onnx>=1.6.0 is required")
return
except:
print("[ERROR] onnx is not installed, use \"pip install onnx==1.6.0\".")
......@@ -240,11 +240,12 @@ def main():
if args.framework == "tensorflow":
assert args.model is not None, "--model should be defined while translating tensorflow model"
without_data_format_optimization = False
assert args.without_data_format_optimization in [
"True", "False"
], "--the param without_data_format_optimization should be defined True or False"
define_input_shape = False
params_merge = False
if args.without_data_format_optimization:
without_data_format_optimization = True
without_data_format_optimization = True if args.without_data_format_optimization == "True" else False
if args.define_input_shape:
define_input_shape = True
if args.params_merge:
......
x2paddle/decoder/caffe.proto 0 → 100644
浏览文件 @ ce6ffee2
syntax = "proto2";
package 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];
// 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];
}
// The label (display) name and label id.
message LabelMapItem {
// Both name and label are required.
optional string name = 1;
optional int32 label = 2;
// display_name is optional.
optional string display_name = 3;
}
message LabelMap {
repeated LabelMapItem item = 1;
}
// Sample a bbox in the normalized space [0, 1] with provided constraints.
message Sampler {
// Minimum scale of the sampled bbox.
optional float min_scale = 1 [default = 1.];
// Maximum scale of the sampled bbox.
optional float max_scale = 2 [default = 1.];
// Minimum aspect ratio of the sampled bbox.
optional float min_aspect_ratio = 3 [default = 1.];
// Maximum aspect ratio of the sampled bbox.
optional float max_aspect_ratio = 4 [default = 1.];
}
// Constraints for selecting sampled bbox.
message SampleConstraint {
// Minimum Jaccard overlap between sampled bbox and all bboxes in
// AnnotationGroup.
optional float min_jaccard_overlap = 1;
// Maximum Jaccard overlap between sampled bbox and all bboxes in
// AnnotationGroup.
optional float max_jaccard_overlap = 2;
// Minimum coverage of sampled bbox by all bboxes in AnnotationGroup.
optional float min_sample_coverage = 3;
// Maximum coverage of sampled bbox by all bboxes in AnnotationGroup.
optional float max_sample_coverage = 4;
// Minimum coverage of all bboxes in AnnotationGroup by sampled bbox.
optional float min_object_coverage = 5;
// Maximum coverage of all bboxes in AnnotationGroup by sampled bbox.
optional float max_object_coverage = 6;
}
// Sample a batch of bboxes with provided constraints.
message BatchSampler {
// Use original image as the source for sampling.
optional bool use_original_image = 1 [default = true];
// Constraints for sampling bbox.
optional Sampler sampler = 2;
// Constraints for determining if a sampled bbox is positive or negative.
optional SampleConstraint sample_constraint = 3;
// If provided, break when found certain number of samples satisfing the
// sample_constraint.
optional uint32 max_sample = 4;
// Maximum number of trials for sampling to avoid infinite loop.
optional uint32 max_trials = 5 [default = 100];
}
// Condition for emitting annotations.
message EmitConstraint {
enum EmitType {
CENTER = 0;
MIN_OVERLAP = 1;
}
optional EmitType emit_type = 1 [default = CENTER];
// If emit_type is MIN_OVERLAP, provide the emit_overlap.
optional float emit_overlap = 2;
}
// The normalized bounding box [0, 1] w.r.t. the input image size.
message NormalizedBBox {
optional float xmin = 1;
optional float ymin = 2;
optional float xmax = 3;
optional float ymax = 4;
optional int32 label = 5;
optional bool difficult = 6;
optional float score = 7;
optional float size = 8;
}
// Annotation for each object instance.
message Annotation {
optional int32 instance_id = 1 [default = 0];
optional NormalizedBBox bbox = 2;
}
// Group of annotations for a particular label.
message AnnotationGroup {
optional int32 group_label = 1;
repeated Annotation annotation = 2;
}
// An extension of Datum which contains "rich" annotations.
message AnnotatedDatum {
enum AnnotationType {
BBOX = 0;
}
optional Datum datum = 1;
// If there are "rich" annotations, specify the type of annotation.
// Currently it only supports bounding box.
// If there are no "rich" annotations, use label in datum instead.
optional AnnotationType type = 2;
// Each group contains annotation for a particular class.
repeated AnnotationGroup annotation_group = 3;
}
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: 44 (last added: plateau_winsize)
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;
// Evaluation type.
optional string eval_type = 41 [default = "classification"];
// ap_version: different ways of computing Average Precision.
// Check https://sanchom.wordpress.com/tag/average-precision/ for details.
// 11point: the 11-point interpolated average precision. Used in VOC2007.
// MaxIntegral: maximally interpolated AP. Used in VOC2012/ILSVRC.
// Integral: the natural integral of the precision-recall curve.
optional string ap_version = 42 [default = "Integral"];
// If true, display per class result.
optional bool show_per_class_result = 44 [default = false];
// 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))))
// - plateau: decreases lr
// if the minimum loss isn't updated for 'plateau_winsize' iters
//
// 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;
// the stepsize for learning rate policy "plateau"
repeated int32 plateau_winsize = 43;
// 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];
}
// 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
optional float minimum_loss = 5 [default = 1E38]; // Historical minimum loss
optional int32 iter_last_event = 6 [default = 0]; // The iteration when last lr-update or min_loss-update happend
}
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: 147 (last added: recurrent_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 AnnotatedDataParameter annotated_data_param = 200;
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 DetectionEvaluateParameter detection_evaluate_param = 205;
optional DetectionOutputParameter detection_output_param = 204;
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 MultiBoxLossParameter multibox_loss_param = 201;
optional MVNParameter mvn_param = 120;
optional NormalizeParameter norm_param = 206;
optional ParameterParameter parameter_param = 145;
optional PermuteParameter permute_param = 202;
optional PoolingParameter pooling_param = 121;
optional PowerParameter power_param = 122;
optional PReLUParameter prelu_param = 131;
optional PriorBoxParameter prior_box_param = 203;
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 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 VideoDataParameter video_data_param = 207;
optional WindowDataParameter window_data_param = 129;
optional AxpyParameter axpy_param = 210;
optional UpsampleParameter upsample_param = 211;
optional ROIPoolingParameter roi_pooling_param = 212;
optional ShuffleChannelParameter shuffle_channel_param = 213;
}
// 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];
optional uint32 crop_h = 11 [default = 0];
optional uint32 crop_w = 12 [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];
// Resize policy
optional ResizeParameter resize_param = 8;
// Noise policy
optional NoiseParameter noise_param = 9;
// Distortion policy
optional DistortionParameter distort_param = 13;
// Expand policy
optional ExpansionParameter expand_param = 14;
// Constraint for emitting the annotation after transformation.
optional EmitConstraint emit_constraint = 10;
}
// Message that stores parameters used by data transformer for resize policy
message ResizeParameter {
//Probability of using this resize policy
optional float prob = 1 [default = 1];
enum Resize_mode {
WARP = 1;
FIT_SMALL_SIZE = 2;
FIT_LARGE_SIZE_AND_PAD = 3;
}
optional Resize_mode resize_mode = 2 [default = WARP];
optional uint32 height = 3 [default = 0];
optional uint32 width = 4 [default = 0];
// A parameter used to update bbox in FIT_SMALL_SIZE mode.
optional uint32 height_scale = 8 [default = 0];
optional uint32 width_scale = 9 [default = 0];
enum Pad_mode {
CONSTANT = 1;
MIRRORED = 2;
REPEAT_NEAREST = 3;
}
// Padding mode for BE_SMALL_SIZE_AND_PAD mode and object centering
optional Pad_mode pad_mode = 5 [default = CONSTANT];
// if specified can be repeated once (would fill all the channels)
// or can be repeated the same number of times as channels
// (would use it them to the corresponding channel)
repeated float pad_value = 6;
enum Interp_mode { //Same as in OpenCV
LINEAR = 1;
AREA = 2;
NEAREST = 3;
CUBIC = 4;
LANCZOS4 = 5;
}
//interpolation for for resizing
repeated Interp_mode interp_mode = 7;
}
message SaltPepperParameter {
//Percentage of pixels
optional float fraction = 1 [default = 0];
repeated float value = 2;
}
// Message that stores parameters used by data transformer for transformation
// policy
message NoiseParameter {
//Probability of using this resize policy
optional float prob = 1 [default = 0];
// Histogram equalized
optional bool hist_eq = 2 [default = false];
// Color inversion
optional bool inverse = 3 [default = false];
// Grayscale
optional bool decolorize = 4 [default = false];
// Gaussian blur
optional bool gauss_blur = 5 [default = false];
// JPEG compression quality (-1 = no compression)
optional float jpeg = 6 [default = -1];
// Posterization
optional bool posterize = 7 [default = false];
// Erosion
optional bool erode = 8 [default = false];
// Salt-and-pepper noise
optional bool saltpepper = 9 [default = false];
optional SaltPepperParameter saltpepper_param = 10;
// Local histogram equalization
optional bool clahe = 11 [default = false];
// Color space conversion
optional bool convert_to_hsv = 12 [default = false];
// Color space conversion
optional bool convert_to_lab = 13 [default = false];
}
// Message that stores parameters used by data transformer for distortion policy
message DistortionParameter {
// The probability of adjusting brightness.
optional float brightness_prob = 1 [default = 0.0];
// Amount to add to the pixel values within [-delta, delta].
// The possible value is within [0, 255]. Recommend 32.
optional float brightness_delta = 2 [default = 0.0];
// The probability of adjusting contrast.
optional float contrast_prob = 3 [default = 0.0];
// Lower bound for random contrast factor. Recommend 0.5.
optional float contrast_lower = 4 [default = 0.0];
// Upper bound for random contrast factor. Recommend 1.5.
optional float contrast_upper = 5 [default = 0.0];
// The probability of adjusting hue.
optional float hue_prob = 6 [default = 0.0];
// Amount to add to the hue channel within [-delta, delta].
// The possible value is within [0, 180]. Recommend 36.
optional float hue_delta = 7 [default = 0.0];
// The probability of adjusting saturation.
optional float saturation_prob = 8 [default = 0.0];
// Lower bound for the random saturation factor. Recommend 0.5.
optional float saturation_lower = 9 [default = 0.0];
// Upper bound for the random saturation factor. Recommend 1.5.
optional float saturation_upper = 10 [default = 0.0];
// The probability of randomly order the image channels.
optional float random_order_prob = 11 [default = 0.0];
}
// Message that stores parameters used by data transformer for expansion policy
message ExpansionParameter {
//Probability of using this expansion policy
optional float prob = 1 [default = 1];
// The ratio to expand the image.
optional float max_expand_ratio = 2 [default = 1.];
}
// 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 AnnotatedDataParameter {
// Define the sampler.
repeated BatchSampler batch_sampler = 1;
// Store label name and label id in LabelMap format.
optional string label_map_file = 2;
// If provided, it will replace the AnnotationType stored in each
// AnnotatedDatum.
optional AnnotatedDatum.AnnotationType anno_type = 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, accumulate global mean/variance values via a moving average. If
// true, use those accumulated values instead of computing mean/variance
// across the batch.
optional bool use_global_stats = 1;
// How much does the moving average decay each iteration?
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;
}
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 à 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 (Number of batches to prefetch to host memory, increase if
// data access bandwidth varies).
optional uint32 prefetch = 10 [default = 4];
}
// Message that store parameters used by DetectionEvaluateLayer
message DetectionEvaluateParameter {
// Number of classes that are actually predicted. Required!
optional uint32 num_classes = 1;
// Label id for background class. Needed for sanity check so that
// background class is neither in the ground truth nor the detections.
optional uint32 background_label_id = 2 [default = 0];
// Threshold for deciding true/false positive.
optional float overlap_threshold = 3 [default = 0.5];
// If true, also consider difficult ground truth for evaluation.
optional bool evaluate_difficult_gt = 4 [default = true];
// 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
// ...
// If provided, we will scale the prediction and ground truth NormalizedBBox
// for evaluation.
optional string name_size_file = 5;
// The resize parameter used in converting NormalizedBBox to original image.
optional ResizeParameter resize_param = 6;
}
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 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 that store parameters used by DetectionOutputLayer
message DetectionOutputParameter {
// Number of classes to be predicted. Required!
optional uint32 num_classes = 1;
// If true, bounding box are shared among different classes.
optional bool share_location = 2 [default = true];
// Background label id. If there is no background class,
// set it as -1.
optional int32 background_label_id = 3 [default = 0];
// Parameters used for non maximum suppression.
optional NonMaximumSuppressionParameter nms_param = 4;
// Parameters used for saving detection results.
optional SaveOutputParameter save_output_param = 5;
// Type of coding method for bbox.
optional PriorBoxParameter.CodeType code_type = 6 [default = CORNER];
// If true, variance is encoded in target; otherwise we need to adjust the
// predicted offset accordingly.
optional bool variance_encoded_in_target = 8 [default = false];
// Number of total bboxes to be kept per image after nms step.
// -1 means keeping all bboxes after nms step.
optional int32 keep_top_k = 7 [default = -1];
// Only consider detections whose confidences are larger than a threshold.
// If not provided, consider all boxes.
optional float confidence_threshold = 9;
// If true, visualize the detection results.
optional bool visualize = 10 [default = false];
// The threshold used to visualize the detection results.
optional float visualize_threshold = 11;
// If provided, save outputs to video file.
optional string save_file = 12;
}
message DropoutParameter {
optional float dropout_ratio = 1 [default = 0.5]; // dropout ratio
}
// 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;
}
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 that store parameters used by MultiBoxLossLayer
message MultiBoxLossParameter {
// Localization loss type.
enum LocLossType {
L2 = 0;
SMOOTH_L1 = 1;
}
optional LocLossType loc_loss_type = 1 [default = SMOOTH_L1];
// Confidence loss type.
enum ConfLossType {
SOFTMAX = 0;
LOGISTIC = 1;
}
optional ConfLossType conf_loss_type = 2 [default = SOFTMAX];
// Weight for localization loss.
optional float loc_weight = 3 [default = 1.0];
// Number of classes to be predicted. Required!
optional uint32 num_classes = 4;
// If true, bounding box are shared among different classes.
optional bool share_location = 5 [default = true];
// Matching method during training.
enum MatchType {
BIPARTITE = 0;
PER_PREDICTION = 1;
}
optional MatchType match_type = 6 [default = PER_PREDICTION];
// If match_type is PER_PREDICTION, use overlap_threshold to
// determine the extra matching bboxes.
optional float overlap_threshold = 7 [default = 0.5];
// Use prior for matching.
optional bool use_prior_for_matching = 8 [default = true];
// Background label id.
optional uint32 background_label_id = 9 [default = 0];
// If true, also consider difficult ground truth.
optional bool use_difficult_gt = 10 [default = true];
// If true, perform negative mining.
// DEPRECATED: use mining_type instead.
optional bool do_neg_mining = 11;
// The negative/positive ratio.
optional float neg_pos_ratio = 12 [default = 3.0];
// The negative overlap upperbound for the unmatched predictions.
optional float neg_overlap = 13 [default = 0.5];
// Type of coding method for bbox.
optional PriorBoxParameter.CodeType code_type = 14 [default = CORNER];
// If true, encode the variance of prior box in the loc loss target instead of
// in bbox.
optional bool encode_variance_in_target = 16 [default = false];
// If true, map all object classes to agnostic class. It is useful for learning
// objectness detector.
optional bool map_object_to_agnostic = 17 [default = false];
// If true, ignore cross boundary bbox during matching.
// Cross boundary bbox is a bbox who is outside of the image region.
optional bool ignore_cross_boundary_bbox = 18 [default = false];
// If true, only backpropagate on corners which are inside of the image
// region when encode_type is CORNER or CORNER_SIZE.
optional bool bp_inside = 19 [default = false];
// Mining type during training.
// NONE : use all negatives.
// MAX_NEGATIVE : select negatives based on the score.
// HARD_EXAMPLE : select hard examples based on "Training Region-based Object Detectors with Online Hard Example Mining", Shrivastava et.al.
enum MiningType {
NONE = 0;
MAX_NEGATIVE = 1;
HARD_EXAMPLE = 2;
}
optional MiningType mining_type = 20 [default = MAX_NEGATIVE];
// Parameters used for non maximum suppression durig hard example mining.
optional NonMaximumSuppressionParameter nms_param = 21;
optional int32 sample_size = 22 [default = 64];
optional bool use_prior_for_nms = 23 [default = false];
}
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 that stores parameters used by NormalizeLayer
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];
}
message ParameterParameter {
optional BlobShape shape = 1;
}
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 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];
}
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 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 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;
}
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 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 VideoDataParameter{
enum VideoType {
WEBCAM = 0;
VIDEO = 1;
}
optional VideoType video_type = 1 [default = WEBCAM];
optional int32 device_id = 2 [default = 0];
optional string video_file = 3;
// Number of frames to be skipped before processing a frame.
optional uint32 skip_frames = 4 [default = 0];
}
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;
}
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 paramters are shared across channels.
optional bool channel_shared = 2 [default = false];
}
message AxpyParameter{
}
message UpsampleParameter{
optional int32 scale = 1 [default = 1];
}
message ROIPoolingParameter {
// 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];
}
message ShuffleChannelParameter {
optional uint32 group = 1[default = 1]; // The number of group
}
x2paddle/decoder/caffe_pb2.py
浏览文件 @ ce6ffee2
因为 它太大了无法显示 source diff 。你可以改为 查看blob。
x2paddle/decoder/onnx_shape_inference.py
浏览文件 @ ce6ffee2
......@@ -267,9 +267,9 @@ class SymbolicShapeInference:
if pending_nodes and self.verbose_ > 0:
print('SymbolicShapeInference: orphaned nodes discarded: ')
print(
* [n.op_type + ': ' + n.output[0] for n in pending_nodes],
sep='\n')
for n in pending_nodes:
print(n.op_type + ': ' + n.output[0])
if input_shapes is not None:
for input_name, shape in input_shapes.items():
for idx in range(len(self.out_mp_.graph.input)):
......
x2paddle/op_mapper/caffe_custom_layer/normalize.py
浏览文件 @ ce6ffee2
......@@ -17,7 +17,7 @@ def normalize_layer(inputs,
scale_param = fluid.layers.create_parameter(
shape=[1] if channel_shared else [1, 1, 1, input_shape[0][1]],
dtype=input.dtype,
attr=name + '_scale')
attr=fluid.ParamAttr(name=name + '_scale'))
scale_param = fluid.layers.reshape(x=scale_param, \
shape=[1] if channel_shared else [input_shape[0][1]])
out = fluid.layers.elementwise_mul(
......
x2paddle/op_mapper/caffe_op_mapper.py
浏览文件 @ ce6ffee2
......@@ -226,7 +226,7 @@ class CaffeOpMapper(OpMapper):
data.append(
np.zeros([output_c, input_c, kernel[0], kernel[1]]).astype(
'float32'))
data.append(np.zeros([output_c, ])).astype('float32')
data.append(np.zeros([output_c, ]).astype('float32'))
else:
data = self.adjust_parameters(node)
self.weights[node.layer_name + '_weights'] = data[0]
......
x2paddle/op_mapper/onnx2paddle/opset9/opset.py
浏览文件 @ ce6ffee2
......@@ -43,6 +43,21 @@ def _const_weight_or_none(node, necessary=False):
return None
def _is_static_shape(shape):
negtive_dims = 0
error_dims = 0
for dim in shape:
if dim < 0:
negtive_dims += 1
if dim < -1:
error_dims += 1
if negtive_dims > 1:
return False
if error_dims > 0:
return False
return True
def _get_same_padding(in_size, kernel_size, stride):
new_size = int(math.ceil(in_size * 1.0 / stride))
pad_size = (new_size - 1) * stride + kernel_size - in_size
......@@ -231,42 +246,9 @@ class OpSet9():
val_x = self.graph.get_input_node(node, idx=0, copy=True)
val_y = self.graph.get_input_node(node, idx=1, copy=True)
val_y_shape = val_y.out_shapes[0]
val_x_shape = val_x.out_shapes[0]
if len(val_x_shape) < len(val_y_shape):
val_x, val_y = val_y, val_x
val_y_shape, val_x_shape = val_x_shape, val_y_shape
str_y_shape = ','.join(str(e) for e in val_y_shape)
str_x_shape = ','.join(str(e) for e in val_x_shape)
slice_idx = 0
if str_y_shape not in str_x_shape:
for dim in val_y_shape:
if dim == 1:
slice_idx += 1
else:
break
attr = {"name": string(node.layer_name)}
if slice_idx < len(val_y_shape) and slice_idx > 0:
val_y_reshaped = val_y_shape[slice_idx:]
var_y_reshaped = val_y.layer_name + '_reshaped'
attr_reshaped = {
'shape': val_y_reshaped,
'name': string(var_y_reshaped)
}
node.fluid_code.add_layer(
'reshape',
inputs=val_y,
output=var_y_reshaped,
param_attr=attr_reshaped)
inputs = {'x': val_x, 'y': var_y_reshaped}
node.fluid_code.add_layer(
op_type, inputs=inputs, output=node, param_attr=attr)
else:
inputs = {'x': val_x, 'y': val_y}
node.fluid_code.add_layer(
op_type, inputs=inputs, output=node, param_attr=attr)
inputs = {'x': val_x, 'y': val_y}
node.fluid_code.add_layer(
op_type, inputs=inputs, output=node, param_attr=None)
@print_mapping_info
def place_holder(self, node):
......@@ -429,6 +411,7 @@ class OpSet9():
output_shape = node.out_shapes[0]
assume_pad2d = False
attr = {}
paddings = []
if len(pads) == 4:
assume_pad2d |= mode != 'constant'
if data_shape:
......@@ -479,8 +462,21 @@ class OpSet9():
output=node,
param_attr={'shape': [1]})
else:
node.fluid_code.add_layer(
'unsqueeze', inputs=val_x, output=node, param_attr=attr)
if str(val_x.dtype) == 'bool':
val_x_cast = val_x.layer_name + '_cast'
node.fluid_code.add_layer(
'cast',
inputs=val_x,
output=val_x_cast,
param_attr={'dtype': string('int64')})
node.fluid_code.add_layer(
'unsqueeze',
inputs=val_x_cast,
output=node,
param_attr=attr)
else:
node.fluid_code.add_layer(
'unsqueeze', inputs=val_x, output=node, param_attr=attr)
@print_mapping_info
def Shrink(self, node):
......@@ -492,16 +488,6 @@ class OpSet9():
node.fluid_code.add_layer(
'hard_shrink', inputs=val_x, output=node, param_attr=attr)
def Greater(self, node):
val_x = self.graph.get_input_node(node, idx=0, copy=True)
val_y = self.graph.get_input_node(node, idx=1, copy=True)
node.fluid_code.add_layer(
'greater_than',
inputs={'x': val_x,
'y': val_y},
output=node,
param_attr=None)
@print_mapping_info
def Constant(self, node):
val_output = self.graph.get_node(node.layer.output[0], copy=True)
......@@ -571,25 +557,26 @@ class OpSet9():
def Expand(self, node):
val_x = self.graph.get_input_node(node, idx=0, copy=True)
val_shape = self.graph.get_input_node(node, idx=1, copy=True)
if len(val_shape.outputs) == 1:
self.omit_nodes.append(val_shape.layer_name)
val_y = self.graph.get_node(node.layer.output[0], copy=True)
out_shape = node.out_shapes[0]
val_x_dtype = val_x.dtype
name_ones = node.layer_name + '_ones'
attr_ones = {'shape': out_shape, 'dtype': string(val_x_dtype)}
attr_ones = {
'shape': val_shape.layer_name,
'dtype': string(val_x_dtype),
'value': 1
}
node.fluid_code.add_layer(
'ones', inputs=None, output=name_ones, param_attr=attr_ones)
'fill_constant',
inputs=None,
output=name_ones,
param_attr=attr_ones)
inputs = {'x': name_ones, 'y': val_x}
attr = {'name': string(node.layer_name)}
node.fluid_code.add_layer(
'elementwise_mul',
inputs=inputs,
output=node.layer_name,
param_attr=attr)
param_attr=None)
@print_mapping_info
def Gather(self, node):
......@@ -600,12 +587,35 @@ class OpSet9():
#assert len(
# indices_shape) <= 2, "Gather op don't support dim of indice >2 "
if axis == 0 and len(indices_shape) <= 1:
node.fluid_code.add_layer(
'gather',
inputs={'input': val_x,
'index': indices},
output=node,
param_attr=None)
if len(val_x.out_shapes[0]) <= 1:
node.fluid_code.add_layer(
'gather',
inputs={'input': val_x,
'index': indices},
output=node,
param_attr=None)
elif len(val_x.out_shapes[0]) > 1:
if len(indices_shape) == 0:
gather_ = node.layer_name + '_1'
node.fluid_code.add_layer(
'gather',
inputs={'input': val_x,
'index': indices},
output=gather_,
param_attr=None)
node.fluid_code.add_layer(
'squeeze',
inputs={'input': gather_,
'axes': [0]},
output=node,
param_attr=None)
else:
node.fluid_code.add_layer(
'gather',
inputs={'input': val_x,
'index': indices},
output=node,
param_attr=None)
elif axis > 0 and len(indices_shape) <= 1:
perm = list(range(len(val_x.out_shapes[0])))
perm = [axis] + perm[:axis] + perm[axis + 1:]
......@@ -624,12 +634,25 @@ class OpSet9():
param_attr=None)
node.fluid_code.add_layer(
'transpose', inputs=node, output=node, param_attr=attr_trans)
if len(indices_shape) < 1:
node.fluid_code.add_layer(
'squeeze',
inputs={'input': node,
'axes': [axis]},
output=node,
param_attr=None)
elif axis == 0 and len(indices_shape) > 1:
if val_x.out_shapes[0] is not None and isinstance(
val_x, ONNXGraphDataNode):
indices_cast = indices.layer_name + '_cast'
node.fluid_code.add_layer(
'embedding',
'cast',
inputs=indices,
output=indices_cast,
param_attr={'dtype': string('int64')})
node.fluid_code.add_layer(
'embedding',
inputs=indices_cast,
output=node,
use_fluid=True,
param_attr={
......@@ -638,7 +661,6 @@ class OpSet9():
})
else:
from functools import reduce
#indices_shape = [1,7]
reshape_shape = reduce(lambda x, y: x * y, indices_shape)
indices_reshape = indices.layer_name + '_shape'
node.fluid_code.add_layer(
......@@ -678,7 +700,7 @@ class OpSet9():
perm = list(range(len(val_x.out_shapes[0])))
perm = [axis] + perm[:axis] + perm[axis + 1:]
attr_trans = {'perm': perm}
name_trans = val_x.layer_name + '_trans'
name_trans = val_x.layer_name + '_transpose'
node.fluid_code.add_layer(
'transpose',
inputs=val_x,
......@@ -690,8 +712,12 @@ class OpSet9():
'index': indices_reshape},
output=node,
param_attr=None)
input_transpose = node.layer_name + '_transpose'
node.fluid_code.add_layer(
'transpose', inputs=node, output=node, param_attr=attr_trans)
'transpose',
inputs=node,
output=input_transpose,
param_attr=attr_trans)
val_x_shape = val_x.out_shapes[0]
reshaped_shape = []
for i in perm:
......@@ -700,10 +726,90 @@ class OpSet9():
reshaped_shape.append(i)
node.fluid_code.add_layer(
'reshape',
inputs=node,
inputs=input_transpose,
output=node,
param_attr={'shape': reshaped_shape})
@print_mapping_info
def ScatterND(self, node):
val_x = self.graph.get_input_node(node, idx=0, copy=True)
indices = self.graph.get_input_node(node, idx=1, copy=True)
updates = self.graph.get_input_node(node, idx=2, copy=True)
if len(indices.out_shapes[0]) == 1:
node.fluid_code.add_layer(
'scatter',
inputs={'input': val_x,
'index': indices,
'updates': updates},
output=node,
param_attr=None)
else:
input_inner_indices = node.layer_name + '_input_inner_indices'
node.fluid_code.add_layer(
'scatter_nd',
inputs={
'shape': val_x.out_shapes[0],
'index': indices,
'updates': updates
},
output=input_inner_indices,
param_attr=None)
constant_minus_one = node.layer_name + '_constant_minus_one'
node.fluid_code.add_layer(
'fill_constant',
inputs=None,
output=constant_minus_one,
param_attr={
'shape': updates.out_shapes[0],
'dtype': string(updates.dtype),
'value': -1
})
indices_mask = node.layer_name + '_indices_mask'
node.fluid_code.add_layer(
'scatter_nd',
inputs={
'shape': val_x.out_shapes[0],
'index': indices,
'updates': constant_minus_one
},
output=indices_mask,
param_attr=None)
constant_1 = node.layer_name + '_constant_1'
node.fluid_code.add_layer(
'fill_constant',
inputs=None,
output=constant_1,
param_attr={
'shape': val_x.out_shapes[0],
'dtype': string(val_x.dtype),
'value': 1
})
input_out_indices_mask = node.layer_name + '_input_out_indices_mask'
node.fluid_code.add_layer(
"elementwise_add",
inputs={"x": indices_mask,
"y": constant_1},
output=input_out_indices_mask,
param_attr=None)
input_out_indices = node.layer_name + '_input_out_indices'
node.fluid_code.add_layer(
"elementwise_mul",
inputs={"x": val_x,
"y": input_out_indices_mask},
output=input_out_indices,
param_attr=None)
node.fluid_code.add_layer(
"elementwise_add",
inputs={"x": input_inner_indices,
"y": input_out_indices},
output=node,
param_attr=None)
@print_mapping_info
def Range(self, node):
val_start = self.graph.get_input_node(node, idx=0, copy=True)
......@@ -754,17 +860,21 @@ class OpSet9():
}
else:
if starts.dtype != 'int32':
starts_cast = starts.layer_name + '_cast'
node.fluid_code.add_layer(
'cast',
inputs=starts,
output=starts,
output=starts_cast,
param_attr={'dtype': string('int32')})
attr['starts'] = starts_cast
if ends.dtype != 'int32':
ends_cast = ends.layer_name + '_cast'
node.fluid_code.add_layer(
'cast',
inputs=ends,
output=ends,
output=ends_cast,
param_attr={'dtype': string('int32')})
attr['ends'] = ends_cast
else:
starts = node.get_attr('starts')
ends = node.get_attr('ends')
......@@ -789,8 +899,6 @@ class OpSet9():
'this is not supported')
if len(value) == 1:
value = value[0]
if dtype.name == 'int64':
dtype = 'int32'
attr = {
'shape': val_shape.layer_name,
'dtype': string(dtype),
......@@ -831,6 +939,14 @@ class OpSet9():
inputs={'x': val_x},
output=node,
param_attr={'shape': shape_value.tolist()})
elif len(node.out_shapes[0]) > 0 and _is_static_shape(node.out_shapes[
0]):
node.fluid_code.add_layer(
'reshape',
inputs={'x': val_x,
'shape': node.out_shapes[0]},
output=node,
param_attr=attr)
elif val_shape.dtype == 'int64':
val_shape_cast = val_shape.layer_name + '_cast'
node.fluid_code.add_layer(
......@@ -882,6 +998,11 @@ class OpSet9():
node.fluid_code.add_layer(
'cast', inputs=val_input, output=node, param_attr=attr)
@print_mapping_info
def Not(self, node):
val_input = self.graph.get_input_node(node, idx=0, copy=True)
node.fluid_code.add_layer('logical_not', inputs=val_input, output=node)
@print_mapping_info
def AveragePool(self, node):
val_x = self.graph.get_input_node(node, idx=0, copy=True)
......@@ -922,12 +1043,16 @@ class OpSet9():
@print_mapping_info
def Concat(self, node):
inputs = []
dtypes = set()
for i in range(len(node.layer.input)):
ipt = self.graph.get_input_node(node, idx=i, copy=True)
if isinstance(ipt, str):
inputs.append(ipt)
else:
inputs.append(ipt.layer_name)
dtypes.add(ipt.dtype)
if len(dtypes) > 1:
assert 'Unspported situation happened, please create issue on https://github.com/PaddlePaddle/X2Paddle/issues.'
axis = node.get_attr('axis')
attr = {'axis': axis}
node.fluid_code.add_layer(
......@@ -1015,10 +1140,22 @@ class OpSet9():
def MatMul(self, node):
val_x = self.graph.get_input_node(node, idx=0, copy=True)
val_y = self.graph.get_input_node(node, idx=1, copy=True)
x_shape = val_x.out_shapes[0]
y_shape = val_y.out_shapes[0]
inputs = {"x": val_x, "y": val_y}
attr = {"name": string(node.layer_name)}
node.fluid_code.add_layer(
"matmul", inputs=inputs, output=node, param_attr=attr)
if y_shape[0] == 1 and x_shape[-1] != 1 and x_shape[0] != 1:
y_squeeze = val_y.layer_name + '_squeeze'
node.fluid_code.add_layer(
"squeeze",
inputs=val_y,
output=y_squeeze,
param_attr={'axes': [0]})
inputs['y'] = y_squeeze
node.fluid_code.add_layer(
"matmul", inputs=inputs, output=node, param_attr=None)
else:
node.fluid_code.add_layer(
"matmul", inputs=inputs, output=node, param_attr=None)
@print_mapping_info
def BatchNormalization(self, node):
......@@ -1154,7 +1291,6 @@ class OpSet9():
'y': cast_condition},
output=mul_val_x,
param_attr=None)
mul_val_y = val_y.layer_name + '_mul'
node.fluid_code.add_layer(
"elementwise_mul",
......@@ -1204,6 +1340,15 @@ class OpSet9():
if repeats is None:
repeats = val_repeats.layer_name
if val_repeats.dtype != 'int32':
attr = {"dtype": string("int32")}
node.fluid_code.add_layer(
"cast",
inputs=repeats,
output="{}.tmp".format(repeats),
param_attr=attr)
repeats = "{}.tmp".format(repeats)
elif isinstance(repeats, int):
repeats = [repeats]
......
x2paddle/op_mapper/paddle2onnx/opset11/opset.py
浏览文件 @ ce6ffee2
......@@ -93,16 +93,13 @@ class OpSet11(OpSet10):
else:
coordinate_transformation_mode = 'half_pixel'
roi_name = self.get_name(op.type, 'roi')
roi_node = self.make_constant_node(roi_name, onnx_pb.TensorProto.FLOAT,
[1, 1, 1, 1, 1, 1, 1, 1])
if ('OutSize' in input_names and len(op.input('OutSize')) > 0) or (
'SizeTensor' in input_names and
len(op.input('SizeTensor')) > 0):
node_list = list()
roi_node = self.make_constant_node(
self.get_name(op.type, 'roi'), onnx_pb.TensorProto.FLOAT,
[1, 1, 1, 1, 1, 1, 1, 1])
roi_name = self.get_name(op.type, 'roi')
roi_node = self.make_constant_node(
roi_name, onnx_pb.TensorProto.FLOAT, [1, 1, 1, 1, 1, 1, 1, 1])
empty_name = self.get_name(op.type, 'empty')
empty_tensor = helper.make_tensor(
empty_name,
......@@ -168,7 +165,7 @@ class OpSet11(OpSet10):
elif 'Scale' in input_names and len(op.input('Scale')) > 0:
node = helper.make_node(
'Resize',
inputs=[op.input('X')[0], op.input('Scale')[0]],
inputs=[op.input('X')[0], roi_name, op.input('Scale')[0]],
outputs=op.output('Out'),
mode='linear',
coordinate_transformation_mode=coordinate_transformation_mode)
......@@ -180,10 +177,6 @@ class OpSet11(OpSet10):
scale_node = self.make_constant_node(scale_name,
onnx_pb.TensorProto.FLOAT,
[1, 1, scale, scale])
roi_name = self.get_name(op.type, 'roi')
roi_node = self.make_constant_node(roi_name,
onnx_pb.TensorProto.FLOAT,
[1, 1, 1, 1, 1, 1, 1, 1])
node = helper.make_node(
'Resize',
inputs=[op.input('X')[0], roi_name, scale_name],
......@@ -194,7 +187,7 @@ class OpSet11(OpSet10):
return [scale_node, roi_node, node]
else:
raise Exception("Unexpected situation happend")
return node
return [roi_node, node]
def nearest_interp(self, op, block):
input_names = op.input_names
......@@ -203,18 +196,21 @@ class OpSet11(OpSet10):
if align_corners:
coordinate_transformation_mode = 'align_corners'
else:
coordinate_transformation_mode = 'asymmetric'
coordinate_transformation_mode = 'half_pixel'
roi_name = self.get_name(op.type, 'roi')
roi_node = self.make_constant_node(roi_name, onnx_pb.TensorProto.FLOAT,
[1, 1, 1, 1, 1, 1, 1, 1])
if 'OutSize' in input_names and len(op.input('OutSize')) > 0:
node = helper.make_node(
'Resize',
inputs=[op.input('X')[0], '', op.input('OutSize')[0]],
inputs=[op.input('X')[0], roi_name, op.input('OutSize')[0]],
outputs=op.output('Out'),
mode='nearest',
coordinate_transformation_mode=coordinate_transformation_mode)
elif 'Scale' in input_names and len(op.input('Scale')) > 0:
node = helper.make_node(
'Resize',
inputs=[op.input('X')[0], op.input('Scale')[0]],
inputs=[op.input('X')[0], roi_name, op.input('Scale')[0]],
outputs=op.output('Out'),
mode='nearest',
coordinate_transformation_mode=coordinate_transformation_mode)
......@@ -226,10 +222,6 @@ class OpSet11(OpSet10):
scale_node = self.make_constant_node(scale_name,
onnx_pb.TensorProto.FLOAT,
[1, 1, scale, scale])
roi_name = self.get_name(op.type, 'roi')
roi_node = self.make_constant_node(roi_name,
onnx_pb.TensorProto.FLOAT,
[1, 1, 1, 1, 1, 1, 1, 1])
node = helper.make_node(
'Resize',
inputs=[op.input('X')[0], roi_name, scale_name],
......@@ -240,7 +232,7 @@ class OpSet11(OpSet10):
return [scale_node, roi_node, node]
else:
raise Exception("Unexpected situation happend")
return node
return [roi_node, node]
def hard_swish(self, op, block):
min_name = self.get_name(op.type, 'min')
......
x2paddle/op_mapper/paddle2onnx/opset11/paddle_custom_layer/multiclass_nms.py
浏览文件 @ ce6ffee2
......@@ -72,6 +72,8 @@ def multiclass_nms(op, block):
dims=(),
vals=[float(attrs['nms_threshold'])]))
boxes_num = block.var(outputs['Out'][0]).shape[0]
top_k_value = np.int64(boxes_num if attrs['keep_top_k'] == -1 else attrs['keep_top_k'])
node_keep_top_k = onnx.helper.make_node(
'Constant',
inputs=[],
......@@ -80,7 +82,7 @@ def multiclass_nms(op, block):
name=name_keep_top_k[0] + "@const",
data_type=onnx.TensorProto.INT64,
dims=(),
vals=[np.int64(attrs['keep_top_k'])]))
vals=[top_k_value]))
node_keep_top_k_2D = onnx.helper.make_node(
'Constant',
......@@ -90,7 +92,7 @@ def multiclass_nms(op, block):
name=name_keep_top_k_2D[0] + "@const",
data_type=onnx.TensorProto.INT64,
dims=[1, 1],
vals=[np.int64(attrs['keep_top_k'])]))
vals=[top_k_value]))
# the paddle data format is x1,y1,x2,y2
kwargs = {'center_point_box': 0}
......
x2paddle/op_mapper/paddle2onnx/opset9/opset.py
浏览文件 @ ce6ffee2
......@@ -174,14 +174,15 @@ class OpSet9(object):
inputs=[op.input('X')[0], temp_value],
outputs=op.output('Out'))
return [shape_node, y_node, node]
elif len(x_shape) == len(y_shape):
elif axis == -1 or axis == (len(x_shape) - 1
) or len(x_shape) == len(y_shape):
node = helper.make_node(
'Add',
inputs=[op.input('X')[0], op.input('Y')[0]],
outputs=op.output('Out'))
return node
else:
raise Excpetion("Unexpected situation happend in elementwise_add")
raise Exception("Unexpected situation happend in elementwise_add")
def elementwise_sub(self, op, block):
axis = op.attr('axis')
......@@ -203,14 +204,15 @@ class OpSet9(object):
inputs=[op.input('X')[0], temp_value],
outputs=op.output('Out'))
return [shape_node, y_node, node]
elif len(x_shape) == len(y_shape):
elif axis == -1 or axis == (len(x_shape) - 1
) or len(x_shape) == len(y_shape):
node = helper.make_node(
'Sub',
inputs=[op.input('X')[0], op.input('Y')[0]],
outputs=op.output('Out'))
return node
else:
raise Excpetion("Unexpected situation happend in elementwise_sub")
raise Exception("Unexpected situation happend in elementwise_sub")
def pool2d(self, op, block):
pool_type = {
......@@ -403,6 +405,22 @@ class OpSet9(object):
'Sum', inputs=op.input('X'), outputs=op.output('Out'))
return node
def floor(self, op, block):
node = helper.make_node(
'Floor', inputs=op.input('X'), outputs=op.output('Out'))
return node
def uniform_random_batch_size_like(self, op, block):
node = helper.make_node(
'RandomUniformLike',
inputs=op.input('Input'),
outputs=op.output('Out'),
high=op.attr('max'),
dtype=self.paddle_onnx_dtype_map[op.attr('dtype')],
low=op.attr('min'),
seed=float(op.attr('seed')), )
return node
def depthwise_conv2d(self, op, block):
return self.conv2d(op, block)
......@@ -444,7 +462,7 @@ class OpSet9(object):
ends = op.attr('ends')
node = helper.make_node(
"Slice",
inputs=[op.input('Input')[0], starts_name, ends_name, axes_name],
inputs=[op.input('Input')[0]],
outputs=op.output('Out'),
axes=axes,
starts=starts,
......@@ -565,7 +583,7 @@ class OpSet9(object):
input_shape = block.vars[op.input('X')[0]].shape
if op.attr('align_corners') or op.attr('align_mode') == 0:
raise Exception(
"Resize in onnx(opset<=10) only support coordinate_transformation_mode: 'asymmetric', Try converting with --onnx_opest 11"
"Resize in onnx(opset<=10) only support coordinate_transformation_mode: 'asymmetric', Try converting with --onnx_opset 11"
)
if ('OutSize' in input_names and len(op.input('OutSize')) > 0) or (
'SizeTensor' in input_names and
......@@ -671,14 +689,82 @@ class OpSet9(object):
input_names = op.input_names
if op.attr('align_corners'):
raise Exception(
"Resize in onnx(opset<=10) only support coordinate_transformation_mode: 'asymmetric', Try converting with --onnx_opest 11"
"Resize in onnx(opset<=10) only support coordinate_transformation_mode: 'asymmetric', Try converting with --onnx_opset 11"
)
if 'OutSize' in input_names and len(op.input('OutSize')) > 0:
node = helper.make_node(
node_list = list()
shape_name0 = self.get_name(op.type, 'shape')
shape_node0 = helper.make_node(
'Shape', inputs=op.input('X'), outputs=[shape_name0])
starts_name = self.get_name(op.type, 'slice.starts')
starts_node = self.make_constant_node(
starts_name, onnx_pb.TensorProto.INT64, [0])
ends_name = self.get_name(op.type, 'slice.ends')
ends_node = self.make_constant_node(ends_name,
onnx_pb.TensorProto.INT64, [2])
shape_name1 = self.get_name(op.type, 'shape')
shape_node1 = helper.make_node(
'Slice',
inputs=[shape_name0, starts_name, ends_name],
outputs=[shape_name1])
node_list.extend([shape_node0, starts_node, ends_node, shape_node1])
if 'OutSize' in input_names and len(op.input('OutSize')) > 0:
cast_shape_name = self.get_name(op.type, "shape.cast")
cast_shape_node = helper.make_node(
'Cast',
inputs=op.input('OutSize'),
outputs=[cast_shape_name],
to=onnx_pb.TensorProto.INT64)
node_list.append(cast_shape_node)
else:
concat_shape_name = self.get_name(
op.type, op.output('Out')[0] + "shape.concat")
concat_shape_node = helper.make_node(
"Concat",
inputs=op.input('SizeTensor'),
outputs=[concat_shape_name],
axis=0)
cast_shape_name = self.get_name(op.type, "shape.cast")
cast_shape_node = helper.make_node(
'Cast',
inputs=[concat_shape_name],
outputs=[cast_shape_name],
to=onnx_pb.TensorProto.INT64)
node_list.extend([concat_shape_node, cast_shape_node])
shape_name2 = self.get_name(op.type, "shape.concat")
shape_node2 = helper.make_node(
'Concat',
inputs=[shape_name1, cast_shape_name],
outputs=[shape_name2],
axis=0)
node_list.append(shape_node2)
cast_shape_name2 = self.get_name(op.type, "shape.cast")
cast_shape_node2 = helper.make_node(
'Cast',
inputs=[shape_name2],
outputs=[cast_shape_name2],
to=onnx_pb.TensorProto.FLOAT)
node_list.append(cast_shape_node2)
cast_shape_name0 = self.get_name(op.type, "shape.cast")
cast_shape_node0 = helper.make_node(
'Cast',
inputs=[shape_name0],
outputs=[cast_shape_name0],
to=onnx_pb.TensorProto.FLOAT)
node_list.append(cast_shape_node0)
outputs_h_w_scales = op.output('Out')[0] + "@out_hw_scales"
node_h_w_scales = helper.make_node(
'Div',
inputs=[cast_shape_name2, cast_shape_name0],
outputs=[outputs_h_w_scales])
node_list.append(node_h_w_scales)
result_node = helper.make_node(
'Resize',
inputs=[op.input('X')[0], op.input('OutSize')[0]],
inputs=[op.input('X')[0], outputs_h_w_scales],
outputs=op.output('Out'),
mode='nearest')
mode='linear')
node_list.extend([result_node])
return node_list
elif 'Scale' in input_names and len(op.input('Scale')) > 0:
node = helper.make_node(
'Resize',
......@@ -714,6 +800,38 @@ class OpSet9(object):
beta=offset)
return node
def swish(self, op, block):
beta = op.attr('beta')
beta_name = self.get_name(op.type, 'beta')
beta_node = onnx.helper.make_node(
'Constant',
name=beta_name,
inputs=[],
outputs=[beta_name],
value=onnx.helper.make_tensor(
name=beta_name,
data_type=onnx.TensorProto.FLOAT,
dims=(),
vals=[beta]))
beta_x_name = self.get_name(op.type, 'beta_x')
beta_x_node = onnx.helper.make_node(
'Mul',
name=beta_x_name,
inputs=[op.input('X')[0], beta_name],
outputs=[beta_x_name])
sigmoid_name = self.get_name(op.type, 'sigmoid')
sigmoid_node = onnx.helper.make_node(
'Sigmoid',
name=sigmoid_name,
inputs=[beta_x_name],
outputs=[sigmoid_name])
swish_node = onnx.helper.make_node(
'Mul',
inputs=[op.input('X')[0], sigmoid_name],
outputs=op.output('Out'))
return [beta_node, beta_x_node, sigmoid_node, swish_node]
def hard_swish(self, op, block):
scale_name = self.get_name(op.type, 'scale')
offset_name = self.get_name(op.type, 'offset')
......@@ -728,8 +846,8 @@ class OpSet9(object):
node0 = helper.make_node(
'Add', inputs=[op.input('X')[0], offset_name], outputs=[name0])
name1 = self.get_name(op.type, 'relu')
min_value = op.attr('min')
max_value = op.attr('max')
min_value = 0.0
max_value = op.attr('threshold')
node1 = helper.make_node(
'Clip',
inputs=[name0],
......@@ -763,14 +881,15 @@ class OpSet9(object):
inputs=[op.input('X')[0], temp_value],
outputs=op.output('Out'))
return [shape_node, y_node, node]
elif len(x_shape) == len(y_shape):
elif axis == -1 or axis == (len(x_shape) - 1
) or len(x_shape) == len(y_shape):
node = helper.make_node(
'Mul',
inputs=[op.input('X')[0], op.input('Y')[0]],
outputs=op.output('Out'))
return node
else:
raise Excpetion("Unexpected situation happend in elementwise_add")
raise Exception("Unexpected situation happend in elementwise_mul")
return node
def feed(self, op, block):
......@@ -799,6 +918,14 @@ class OpSet9(object):
axes=op.attr('axes'))
return node
def cast(self, op, block):
node = helper.make_node(
'Cast',
inputs=op.input('X'),
outputs=op.output('Out'),
to=self.paddle_onnx_dtype_map[op.attr('out_dtype')])
return node
def arg_max(self, op, block):
node = helper.make_node(
'ArgMax',
......
x2paddle/op_mapper/paddle2onnx/opset9/paddle_custom_layer/multiclass_nms.py
浏览文件 @ ce6ffee2
......@@ -72,6 +72,8 @@ def multiclass_nms(op, block):
dims=(),
vals=[float(attrs['nms_threshold'])]))
boxes_num = block.var(outputs['Out'][0]).shape[0]
top_k_value = np.int64(boxes_num if attrs['keep_top_k'] == -1 else attrs['keep_top_k'])
node_keep_top_k = onnx.helper.make_node(
'Constant',
inputs=[],
......@@ -80,7 +82,7 @@ def multiclass_nms(op, block):
name=name_keep_top_k[0] + "@const",
data_type=onnx.TensorProto.INT64,
dims=(),
vals=[np.int64(attrs['keep_top_k'])]))
vals=[top_k_value]))
node_keep_top_k_2D = onnx.helper.make_node(
'Constant',
......@@ -90,7 +92,7 @@ def multiclass_nms(op, block):
name=name_keep_top_k_2D[0] + "@const",
data_type=onnx.TensorProto.INT64,
dims=[1, 1],
vals=[np.int64(attrs['keep_top_k'])]))
vals=[top_k_value]))
# the paddle data format is x1,y1,x2,y2
kwargs = {'center_point_box': 0}
......
x2paddle/op_mapper/tf_op_mapper_nhwc.py
浏览文件 @ ce6ffee2
......@@ -299,6 +299,10 @@ class TFOpMapperNHWC(OpMapper):
data_format = node.get_attr("data_format").decode()
pad_mode = node.get_attr("padding").decode()
channel_first = data_format == "NCHW"
if data_format == "NHWC":
n, h, w, c = input.out_shapes[0]
else:
n, c, h, w = input.out_shapes[0]
if kernel.layer_type == 'Const':
kernel_value = kernel.value
......@@ -329,10 +333,15 @@ class TFOpMapperNHWC(OpMapper):
"dilation": dilations[2:4],
"padding": string(pad_mode)
}
if hasattr(node, 'dilation') and attr['dilation'] == [1, 1]:
if len(node.dilation) == 1:
attr['dilation'] = [1, node.dilation[0]]
if c == -1:
reshape_attr = {"shape": [0, k_size[2], 0, 0]}
node.fluid_code.add_layer(
"reshape", inputs=input, output=input, param_attr=reshape_attr)
node.fluid_code.add_layer(
"conv2d", inputs=input, output=node, param_attr=attr)
if not channel_first:
......@@ -748,11 +757,12 @@ class TFOpMapperNHWC(OpMapper):
self.add_omit_nodes(begin.layer_name, node.layer_name)
begin = begin.value.tolist()
else:
begin = begin
shape = begin.out_shapes[0]
attr = {"shape": shape}
node.fluid_code.add_layer(
"reshape", inputs=begin, output=begin, param_attr=attr)
begin = self.decoder.infer_tensor(begin).tolist()
# shape = begin.out_shapes[0]
# attr = {"shape": shape}
# node.fluid_code.add_layer(
# "reshape", inputs=begin, output=begin, param_attr=attr)
if size.layer_type == "Const":
self.add_omit_nodes(size.layer_name, node.layer_name)
size = size.value.tolist()
......@@ -1058,13 +1068,25 @@ class TFOpMapperNHWC(OpMapper):
axis = axis.value.tolist()
assert axis == 0, "Only support axis=0 in GatherV2 OP"
attr = {'overwrite': False}
embeddings_shape = embeddings.out_shapes[0][-1]
reshape_list = list()
reshape_name = index.layer_name
if len(index.out_shapes[0]) != 1:
reshape_list = index.out_shapes[0]
reshape_attr = {"shape": [-1]}
reshape_name = "{}_reshape".format(index.layer_name)
node.fluid_code.add_layer(
"reshape", inputs=index, output=index, param_attr=reshape_attr)
inputs = {'input': embeddings, 'index': index}
"reshape",
inputs=index,
output=reshape_name,
param_attr=reshape_attr)
inputs = {'input': embeddings, 'index': reshape_name}
node.fluid_code.add_layer(
"gather", inputs=inputs, output=node, param_attr=attr)
if len(index.out_shapes[0]) != 1:
reshape_attr = {"shape": reshape_list + [embeddings_shape]}
node.fluid_code.add_layer(
"reshape", inputs=node, output=node, param_attr=reshape_attr)
def OneShotIterator(self, node):
return self.Placeholder(node)
......
x2paddle/optimizer/tf_optimizer.py
浏览文件 @ ce6ffee2
......@@ -863,6 +863,9 @@ class TFOptimizer(object):
weight = numpy.expand_dims(weight, 2)
weight = numpy.expand_dims(weight, 3)
self.op_mapper.weights[in_nodes3[0].layer_name] = weight
# fix bug in Paddle1.8.3 and may change in next version.
# self.op_mapper.weights[in_nodes3[0].layer_name +
# '_1'] = weight.reshape(1, -1)
in_nodes3[0].fluid_code.layers[0].param_attr["shape"] = [
1, in_shape[-1], 1, 1
]
......
x2paddle_model_zoo.md
浏览文件 @ ce6ffee2
# X2Paddle模型测试库
> 目前X2Paddle支持50+的TensorFlow OP，40+的Caffe Layer，覆盖了大部分CV分类模型常用的操作。我们在如下模型列表中测试了X2Paddle的转换。
> 目前X2Paddle支持70+的TensorFlow OP，40+的Caffe Layer，覆盖了大部分CV分类模型常用的操作。我们在如下模型列表中测试了X2Paddle的转换。
**注：** 受限于不同框架的差异，部分模型可能会存在目前无法转换的情况，如TensorFlow中包含控制流的模型，NLP模型等。对于CV常见的模型，如若您发现无法转换或转换失败，存在较大diff等问题，欢迎通过[ISSUE反馈](https://github.com/PaddlePaddle/X2Paddle/issues/new)的方式告知我们(模型名，代码实现或模型获取方式)，我们会及时跟进：）
......@@ -20,10 +20,13 @@
| ResNet_V1_101 | [code](https://github.com/tensorflow/models/tree/master/research/slim/nets) |-|
| ResNet_V2_101 | [code](https://github.com/tensorflow/models/tree/master/research/slim/nets) |-|
| UNet | [code1](https://github.com/jakeret/tf_unet )/[code2](https://github.com/lyatdawn/Unet-Tensorflow) |-|
|MTCNN | [code](https://github.com/AITTSMD/MTCNN-Tensorflow) |-|
|YOLO-V3| [code](https://github.com/YunYang1994/tensorflow-yolov3) | 转换需要关闭NHWC->NCHW的优化，见[文档Q2](FAQ.md) |
| FALSR | [code](https://github.com/xiaomi-automl/FALSR) | - |
| DCSCN | [code](https://modelzoo.co/model/dcscn-super-resolution) | - |
| MTCNN | [code](https://github.com/AITTSMD/MTCNN-Tensorflow) |-|
| YOLO-V3| [code](https://github.com/YunYang1994/tensorflow-yolov3) | 转换需要关闭NHWC->NCHW的优化，见[文档Q2](FAQ.md) |
| FALSR | [code](https://github.com/xiaomi-automl/FALSR) | 需使用参数without_data_format_optimization |
| DCSCN | [code](https://modelzoo.co/model/dcscn-super-resolution) | 需使用参数without_data_format_optimization |
| Bert（albert） | [code](https://github.com/google-research/albert#pre-trained-models) | 需使用参数without_data_format_optimization |
| Bert（chinese_L-12_H-768_A-12） | [code](https://github.com/google-research/bert#pre-trained-models) | 需使用参数without_data_format_optimization |
| Bert（multi_cased_L-12_H-768_A-12） | [code](https://github.com/google-research/bert#pre-trained-models) | 需使用参数without_data_format_optimization |
## Caffe
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或