def extract(cls, node):
attrs = get_mxnet_layer_attrs(node.symbol_dict)
num_classes = 21
top_k = attrs.int("nms_topk", -1)
keep_top_k = top_k
variance_encoded_in_target = 0
code_type = "caffe.PriorBoxParameter.CENTER_SIZE"
share_location = 1
nms_threshold = attrs.float("nms_threshold", 0.5)
confidence_threshold = attrs.float("threshold", 0.01)
background_label_id = 0
clip = 0 if not attrs.bool("clip", True) else 1
node_attrs = {
'type': 'DetectionOutput',
'op': __class__.op,
'num_classes': num_classes,
'keep_top_k': keep_top_k,
'variance_encoded_in_target': variance_encoded_in_target,
'code_type': code_type,
'share_location': share_location,
'confidence_threshold': confidence_threshold,
'background_label_id': background_label_id,
'nms_threshold': nms_threshold,
'top_k': top_k,
'decrease_label_id': 1,
'clip_before_nms': clip,
'normalized': 1,
}
DetectionOutput.update_node_stat(node, node_attrs)
return cls.enabled
def extract(cls, node):
attrs = {
'variance_encoded_in_target':
int(node.module.variance_encoded_in_target),
'nms_threshold':
node.module.nms_threshold,
'confidence_threshold':
node.module.confidence_threshold,
'top_k':
node.module.top_k,
'keep_top_k':
node.module.keep_top_k,
'code_type':
node.module.code_type,
}
DetectionOutput.update_node_stat(node, attrs)
return cls.enabled
def extract(cls, node):
attrs = get_mxnet_layer_attrs(node.symbol_dict)
# We can not get num_classes attribute from the operation, so it must be set to None.
# In this case num_classes attribute will be defined in the infer function in
# mo/front/common/partial_infer/multi_box_detection.py
num_classes = None
top_k = attrs.int("nms_topk", -1)
keep_top_k = top_k
variance_encoded_in_target = 0
code_type = "caffe.PriorBoxParameter.CENTER_SIZE"
share_location = 1
nms_threshold = attrs.float("nms_threshold", 0.5)
confidence_threshold = attrs.float("threshold", 0.01)
background_label_id = 0
clip = 0 if not attrs.bool("clip", True) else 1
node_attrs = {
'type': 'DetectionOutput',
'op': __class__.op,
'num_classes': num_classes,
'keep_top_k': keep_top_k,
'variance_encoded_in_target': variance_encoded_in_target,
'code_type': code_type,
'share_location': share_location,
'confidence_threshold': confidence_threshold,
'background_label_id': background_label_id,
'nms_threshold': nms_threshold,
'top_k': top_k,
'decrease_label_id': 1,
'clip_before_nms': clip,
'normalized': 1,
}
DetectionOutput.update_node_stat(node, node_attrs)
return cls.enabled
def extract(cls, node):
nms_threshold = onnx_attr(node, 'nms_threshold', 'f', default=0.0)
eta = onnx_attr(node, 'eta', 'f', default=0.0)
top_k = onnx_attr(node, 'top_k', 'i', default=-1)
code_type_values = {
b"CORNER": "caffe.PriorBoxParameter.CORNER",
b"CENTER_SIZE": "caffe.PriorBoxParameter.CENTER_SIZE",
}
code_type = onnx_attr(node,
'code_type',
's',
default=code_type_values[b"CORNER"])
try:
code_type = code_type_values[code_type]
except KeyError:
raise Error(
"Incorrect value of code_type parameter {}".format(code_type))
resize_mode_values = {
b"":
"",
b"WARP":
"caffe.ResizeParameter.WARP",
b"FIT_SMALL_SIZE":
"caffe.ResizeParameter.FIT_SMALL_SIZE",
b"FIT_LARGE_SIZE_AND_PAD":
"caffe.ResizeParameter.FIT_LARGE_SIZE_AND_PAD",
}
resize_mode = onnx_attr(node, 'resize_mode', 's', default=b"")
try:
resize_mode = resize_mode_values[resize_mode]
except KeyError:
raise Error("Incorrect value of resize_mode parameter {}".format(
resize_mode))
pad_mode_values = {
b"": "",
b"CONSTANT": "caffe.ResizeParameter.CONSTANT",
b"MIRRORED": "caffe.ResizeParameter.MIRRORED",
b"REPEAT_NEAREST": "caffe.ResizeParameter.REPEAT_NEAREST"
}
pad_mode = onnx_attr(node, 'pad_mode', 's', default=b"")
try:
pad_mode = pad_mode_values[pad_mode]
except KeyError:
raise Error(
"Incorrect value of pad_mode parameter {}".format(pad_mode))
interp_mode_values = {
b"": "",
b"LINEAR": "caffe.ResizeParameter.LINEAR",
b"AREA": "caffe.ResizeParameter.AREA",
b"NEAREST": "caffe.ResizeParameter.NEAREST",
b"CUBIC": "caffe.ResizeParameter.CUBIC",
b"LANCZOS4": "caffe.ResizeParameter.LANCZOS4"
}
interp_mode = onnx_attr(node, 'interp_mode', 's', default=b"")
try:
interp_mode = interp_mode_values[interp_mode]
except KeyError:
raise Error("Incorrect value of interp_mode parameter {}".format(
interp_mode))
attrs = {
'num_classes':
onnx_attr(node, 'num_classes', 'i', default=0),
'share_location':
onnx_attr(node, 'share_location', 'i', default=0),
'background_label_id':
onnx_attr(node, 'background_label_id', 'i', default=0),
'code_type':
code_type,
'variance_encoded_in_target':
onnx_attr(node, 'variance_encoded_in_target', 'i', default=0),
'keep_top_k':
onnx_attr(node, 'keep_top_k', 'i', default=0),
'confidence_threshold':
onnx_attr(node, 'confidence_threshold', 'f', default=0),
'visualize_threshold':
onnx_attr(node, 'visualize_threshold', 'f', default=0.6),
# nms_param
'nms_threshold':
nms_threshold,
'top_k':
top_k,
'eta':
eta,
# save_output_param.resize_param
'prob':
onnx_attr(node, 'prob', 'f', default=0),
'resize_mode':
resize_mode,
'height':
onnx_attr(node, 'height', 'i', default=0),
'width':
onnx_attr(node, 'width', 'i', default=0),
'height_scale':
onnx_attr(node, 'height_scale', 'i', default=0),
'width_scale':
onnx_attr(node, 'width_scale', 'i', default=0),
'pad_mode':
pad_mode,
'pad_value':
onnx_attr(node, 'pad_value', 's', default=""),
'interp_mode':
interp_mode,
'input_width':
onnx_attr(node, 'input_width', 'i', default=1),
'input_height':
onnx_attr(node, 'input_height', 'i', default=1),
'normalized':
onnx_attr(node, 'normalized', 'i', default=1),
}
# update the attributes of the node
DetectionOutput.update_node_stat(node, attrs)
return cls.enabled
def extract(cls, node):
pl = node.pb
assert pl, 'Protobuf layer can not be empty'
param = pl.detection_output_param
# TODO rewrite params as complex structures
if hasattr(param, 'nms_param'):
nms_threshold = param.nms_param.nms_threshold
eta = param.nms_param.eta
if param.nms_param.top_k == 0:
top_k = -1
else:
top_k = param.nms_param.top_k
code_type_values = [
"", "caffe.PriorBoxParameter.CORNER",
"caffe.PriorBoxParameter.CENTER_SIZE",
"caffe.PriorBoxParameter.CORNER_SIZE"
]
code_type = code_type_values[1]
if hasattr(param, 'code_type'):
if param.code_type < 1 or param.code_type > 3:
log.error("Incorrect value of code_type parameter")
return
code_type = code_type_values[param.code_type]
visualize_threshold = param.visualize_threshold if param.visualize_threshold else 0.6
resize_mode_values = [
"", "caffe.ResizeParameter.WARP",
"caffe.ResizeParameter.FIT_SMALL_SIZE",
"caffe.ResizeParameter.FIT_LARGE_SIZE_AND_PAD"
]
if param.save_output_param.resize_param.resize_mode < 1 or param.save_output_param.resize_param.resize_mode > 3:
log.error("Incorrect value of resize_mode parameter")
return
resize_mode = resize_mode_values[
param.save_output_param.resize_param.resize_mode]
pad_mode_values = [
"", "caffe.ResizeParameter.CONSTANT",
"caffe.ResizeParameter.MIRRORED",
"caffe.ResizeParameter.REPEAT_NEAREST"
]
if param.save_output_param.resize_param.pad_mode < 1 or param.save_output_param.resize_param.pad_mode > 3:
log.error("Incorrect value of pad_mode parameter")
else:
pad_mode = pad_mode_values[
param.save_output_param.resize_param.pad_mode]
interp_mode_values = [
"", "caffe.ResizeParameter.LINEAR", "caffe.ResizeParameter.AREA",
"caffe.ResizeParameter.NEAREST", "caffe.ResizeParameter.CUBIC",
"caffe.ResizeParameter.LANCZOS4"
]
interp_mode = ""
for x in param.save_output_param.resize_param.interp_mode:
if x < 1 or x > 5:
log.error("Incorrect value of interp_mode parameter")
return
interp_mode += interp_mode_values[x]
attrs = {
'num_classes':
param.num_classes,
'share_location':
int(param.share_location),
'background_label_id':
param.background_label_id,
'code_type':
code_type,
'variance_encoded_in_target':
int(param.variance_encoded_in_target),
'keep_top_k':
param.keep_top_k,
'confidence_threshold':
param.confidence_threshold,
'visualize':
param.visualize,
'visualize_threshold':
visualize_threshold,
'save_file':
param.save_file,
# nms_param
'nms_threshold':
nms_threshold,
'top_k':
top_k,
'eta':
eta,
# save_output_param
'output_directory':
param.save_output_param.output_directory,
'output_name_prefix':
param.save_output_param.output_name_prefix,
'output_format':
param.save_output_param.output_format,
'label_map_file':
param.save_output_param.label_map_file,
'name_size_file':
param.save_output_param.name_size_file,
'num_test_image':
param.save_output_param.num_test_image,
# save_output_param.resize_param
'prob':
param.save_output_param.resize_param.prob,
'resize_mode':
resize_mode,
'height':
param.save_output_param.resize_param.height,
'width':
param.save_output_param.resize_param.width,
'height_scale':
param.save_output_param.resize_param.height_scale,
'width_scale':
param.save_output_param.resize_param.width_scale,
'pad_mode':
pad_mode,
'pad_value':
','.join(
str(x)
for x in param.save_output_param.resize_param.pad_value),
'interp_mode':
interp_mode,
}
# these params can be omitted in caffe.proto and in param as consequence,
# so check if it is set or set to default
fields = [field[0].name for field in param.ListFields()]
if 'input_width' in fields:
attrs['input_width'] = param.input_width
if 'input_height' in fields:
attrs['input_height'] = param.input_height
if 'normalized' in fields:
attrs['normalized'] = int(param.normalized)
if 'objectness_score' in fields:
attrs['objectness_score'] = param.objectness_score
mapping_rule = merge_attrs(param, attrs)
# force setting infer function because it doesn't exist in proto so merge_attrs will not set it
mapping_rule.update({'infer': multi_box_detection_infer})
# update the attributes of the node
DetectionOutput.update_node_stat(node, mapping_rule)
return cls.enabled
