version 2.0.2

Author: asiryan

caffe2onnx/proto/__init__.py

@@ -1,7 +0,0 @@
-import os
-import importlib
-
-
-def import_caffe_pb2(caffe_proto_name):
-    caffe_pb2 = importlib.import_module("proto.%s_pb2"%caffe_proto_name)
-    return caffe_pb2

caffe2onnx/proto/caffe.proto

@@ -422,6 +422,20 @@ message LayerParameter {
   optional ThresholdParameter threshold_param = 128;
   optional TileParameter tile_param = 138;
   optional WindowDataParameter window_data_param = 129;
+  optional InterpParameter interp_param = 166;
+  optional ShuffleChannelParameter shuffle_channel_param = 164;
+  optional PermuteParameter permute_param = 202;
+  optional PriorBoxParameter prior_box_param = 203;
+  optional DetectionOutputParameter detection_output_param = 204;
+  optional DetectionEvaluateParameter detection_evaluate_param = 205;
+  optional NormalizeParameter norm_param = 206;
+  optional AxpyParameter axpy_param = 151;
+  optional ReLU6Parameter relu6_param = 100000;
+}
+
+
+message ShuffleChannelParameter {
+  optional uint32 group = 1[default = 1]; // The number of group
 }
 
 // Message that stores parameters used to apply transformation
@@ -1456,3 +1470,207 @@ message PReLUParameter {
   // Whether or not slope parameters are shared across channels.
   optional bool channel_shared = 2 [default = false];
 }
+
+message ReLU6Parameter {
+  optional float negative_slope = 1 [default = 0];
+}
+
+message InterpParameter {
+   optional int32 height = 1 [default = 0]; // Height of output
+   optional int32 width = 2 [default = 0]; // Width of output
+   optional int32 zoom_factor = 3 [default = 1]; // zoom factor
+   optional int32 shrink_factor = 4 [default = 1]; // shrink factor
+   optional int32 pad_beg = 5 [default = 0]; // padding at begin of input
+   optional int32 pad_end = 6 [default = 0]; // padding at end of input
+}
+
+message 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 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;//对应论文2.2节中公式（4）中的sk×网络输入层输入图像[data层的输入]大小
+  // 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; //  等宽比
+  // 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]; // 是否反转等宽比
+  // If true, will clip the prior so that it is within [0, 1]
+  optional bool clip = 5 [default = false]; // 是否进行裁剪,是否保证默认框整个在网络输入层输入图像内）
+  // 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;//网络输入层输入图像的高（或自行设置的高度）
+  optional uint32 img_w = 9;//网络输入层输入图像的宽（或自行设置的高度）
+
+  // Explicitly provide the step size.
+  optional float step = 10;
+  // Either step or step_h/step_w should be specified; not both.
+  optional float step_h = 11;
+  optional float step_w = 12;
+
+  // Offset to the top left corner of each cell.
+  optional float offset = 13 [default = 0.5];
+}
+
+// Message that 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 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 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 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 AxpyParameter {
+
+}
+
+// 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];
+}

caffe2onnx/proto/caffe_pb2.py

@@ -0,0 +1,14 @@
+import caffe2onnx.src.c2oObject as Node
+
+
+def get_add_output_shape(input_shape):
+    output_shape = input_shape[0]
+    return [output_shape]
+
+
+def create_add_node(layer, node_name, input_name, output_name, input_shape):
+    output_shape = get_add_output_shape(input_shape)
+
+    node = Node.c2oNode(layer, node_name, 'Add', input_name, output_name, input_shape, output_shape)
+
+    return node

caffe2onnx/proto/caffe_upsample.proto

@@ -0,0 +1,63 @@
+import caffe2onnx.src.c2oObject as Node
+from typing import *
+import copy
+
+
+def need_add_reshape(input_shape: List[List]) -> bool:
+    return len(input_shape[0]) != len(input_shape[1])
+
+
+def get_param_shape(input_shape: List[List]) -> List:
+    input = input_shape[0]
+    scale = copy.deepcopy(input_shape[1])
+    if len(input) > len(scale):
+        for i in range(len(input) - len(scale)):
+            scale.append(1)
+    return scale
+
+def broadcast_scale(input_shape: List[List]) -> List[List]:
+    input = input_shape[0]
+    scale = input_shape[1]
+    if len(input) > len(scale):
+        for i in range(len(input) - len(scale)):
+            scale.append(1)
+        broadcast_shape = [input, scale]
+    elif len(input) < len(scale):
+        print("the scale should be less than input")
+        exit(-1)
+    else:
+        broadcast_shape = [input, scale]
+    return broadcast_shape
+
+
+def get_mul_output_shape(input_shape: List[List]) -> List[List]:
+    output_shape = input_shape[1]
+    return [output_shape]
+
+
+def create_axpy_mul_node(layer, node_name, input_name, output_name, input_shape):
+
+    new_node_name = node_name + "_middle"
+    output_shape = get_mul_output_shape(input_shape)
+    new_input_name = [input_name[0], input_name[1]]
+    new_output_name = [output_name[0] + "_mul"]
+    new_input_shape = [input_shape[0], input_shape[1]]
+
+    node = Node.c2oNode(layer, new_node_name, 'Mul', new_input_name, new_output_name, new_input_shape, output_shape)
+
+    return node
+
+def get_add_output_shape(input_shape):
+
+    output_shape = input_shape[1]
+
+    return [output_shape]
+
+def create_axpy_add_node(layer, node_name, input_name, output_name, input_shape):
+
+    output_shape = get_add_output_shape(input_shape)
+    new_input_name = [node_name + "_mul", input_name[2]]
+    new_input_shape = [input_shape[1], input_shape[2]]
+    node = Node.c2oNode(layer, node_name, "Add", new_input_name, output_name, input_shape, output_shape)
+
+    return node

caffe2onnx/proto/caffe_upsample_pb2.py

@@ -1,26 +1,22 @@
 import caffe2onnx.src.c2oObject as Node
 
-##-----------------------------BatchNormalization layer = BatchNorm + Scale-------------------------------------##
-# Get hyperparameters
+
 def getBNAttri(layer):
-    # Hyperparameter dictionary
-    
-    dict = {"epsilon": layer.batch_norm_param.eps,  # Slip coefficient
-            "momentum": layer.batch_norm_param.moving_average_fraction
+    eps = layer.batch_norm_param.eps
+    momentum = layer.batch_norm_param.moving_average_fraction
+    dict = {"epsilon": eps,
+            "momentum": momentum
             }
     return dict
 
-# Calculate the output dimension
+
 def getBNOutShape(input_shape):
     output_shape = input_shape
     return output_shape
 
-# Build node
+
 def createBN(layer, nodename, inname, outname, input_shape):
     dict = getBNAttri(layer)
-    # Calculate output_shape, the output dimension is equal to the input dimension
     output_shape = getBNOutShape(input_shape)
-
-    # Build node
     node = Node.c2oNode(layer, nodename, "BatchNormalization", inname, outname, input_shape, output_shape,dict)
-    return node
+    return node

caffe2onnx/src/OPs/Add.py

@@ -0,0 +1,27 @@
+import caffe2onnx.src.c2oObject as Node
+
+
+def get_attribute(layer):
+    attributes = {}
+    max_attribute = 0
+    min_attribute = 0
+    if layer.type == 'ReLU6':
+        max_attribute = 6.0
+        min_attribute = 0
+
+    attribute = {
+        'max': max_attribute,
+        'min': min_attribute
+    }
+    return attributes
+
+
+def get_clip_output_shape(input_shape):
+    output_shape = input_shape
+    return output_shape
+
+
+def create_clip_node(layer, node_name, input_name, output_name, input_shape):
+    output_shape = get_clip_output_shape(input_shape)
+    node = Node.c2oNode(layer, node_name, 'Clip', input_name, output_name, input_shape, output_shape)
+    return node