MyCaffe
1.12.2.41
Deep learning software for Windows C# programmers.
|
Specifies the parameters for the MultiBoxLossParameter. More...
Public Types | |
enum | LocLossType { L2 , SMOOTH_L1 } |
Defines the localization loss types. More... | |
enum | ConfLossType { SOFTMAX , LOGISTIC } |
Defines the confidence loss types. More... | |
enum | MatchType { BIPARTITE , PER_PREDICTION } |
Defines the matching method used during training. More... | |
enum | MiningType { NONE , MAX_NEGATIVE , HARD_EXAMPLE } |
Defines the mining type used during training. More... | |
Public Types inherited from MyCaffe.param.LayerParameterBase | |
enum | LABEL_TYPE { NONE , SINGLE , MULTIPLE , ONLY_ONE } |
Defines the label type. More... | |
Public Member Functions | |
MultiBoxLossParameter () | |
The constructor. More... | |
override object | Load (BinaryReader br, bool bNewInstance=true) |
Load the parameter from a binary reader. More... | |
override void | Copy (LayerParameterBase src) |
Copy on parameter to another. More... | |
override LayerParameterBase | Clone () |
Creates a new copy of this instance of the parameter. More... | |
override RawProto | ToProto (string strName) |
Convert the parameter into a RawProto. More... | |
Public Member Functions inherited from MyCaffe.param.LayerParameterBase | |
LayerParameterBase () | |
Constructor for the parameter. More... | |
virtual string | PrepareRunModelInputs () |
This method gives derivative classes a chance specify model inputs required by the run model. More... | |
virtual void | PrepareRunModel (LayerParameter p) |
This method gives derivative classes a chance to prepare the layer for a run-model. More... | |
void | Save (BinaryWriter bw) |
Save this parameter to a binary writer. More... | |
Public Member Functions inherited from MyCaffe.basecode.BaseParameter | |
BaseParameter () | |
Constructor for the parameter. More... | |
virtual bool | Compare (BaseParameter p) |
Compare this parameter to another parameter. More... | |
Static Public Member Functions | |
static LocLossType | LocLossTypeFromString (string str) |
Convert a string into a LocLossType. More... | |
static ConfLossType | ConfLossTypeFromString (string str) |
Convert a string into a ConfLossType. More... | |
static MatchType | MatchTypeFromString (string str) |
Convert a string into a MatchType. More... | |
static MiningType | MiningTypeFromString (string str) |
Convert a string into a MiningType. More... | |
static MultiBoxLossParameter | FromProto (RawProto rp) |
Parses the parameter from a RawProto. More... | |
Static Public Member Functions inherited from MyCaffe.basecode.BaseParameter | |
static double | ParseDouble (string strVal) |
Parse double values using the US culture if the decimal separator = '.', then using the native culture, and if then lastly trying the US culture to handle prototypes containing '.' as the separator, yet parsed in a culture that does not use '.' as a decimal. More... | |
static bool | TryParse (string strVal, out double df) |
Parse double values using the US culture if the decimal separator = '.', then using the native culture, and if then lastly trying the US culture to handle prototypes containing '.' as the separator, yet parsed in a culture that does not use '.' as a decimal. More... | |
static float | ParseFloat (string strVal) |
Parse float values using the US culture if the decimal separator = '.', then using the native culture, and if then lastly trying the US culture to handle prototypes containing '.' as the separator, yet parsed in a culture that does not use '.' as a decimal. More... | |
static bool | TryParse (string strVal, out float f) |
Parse doufloatble values using the US culture if the decimal separator = '.', then using the native culture, and if then lastly trying the US culture to handle prototypes containing '.' as the separator, yet parsed in a culture that does not use '.' as a decimal. More... | |
Properties | |
LocLossType | loc_loss_type [getset] |
Get/set the localization loss type (default = SMOOTH_L1). More... | |
ConfLossType | conf_loss_type [getset] |
Get/set the confidence loss type (default = SOFTMAX). More... | |
float | loc_weight [getset] |
Get/set the weight for the localization loss (default = 1.0). More... | |
uint | num_classes [getset] |
Get/set the number of classes to be predicted - required! More... | |
bool | share_location [getset] |
Get/sets whether or not the bounding box is shared among different classes (default = true). More... | |
MatchType | match_type [getset] |
Get/set the matching method used during training (default = PER_PREDICTION). More... | |
float | overlap_threshold [getset] |
Get/set the overlap threshold (default = 0.5). More... | |
uint | background_label_id [getset] |
Get/set the background label id. More... | |
bool | use_difficult_gt [getset] |
Get/set whether or not to consider the difficult ground truth (defalt = true). More... | |
bool? | do_neg_mining [getset] |
DEPRECIATED: Get/set whether or not to perform negative mining (default = false). More... | |
float | neg_pos_ratio [getset] |
Get/set the negative/positive ratio (default = 3.0). More... | |
float | neg_overlap [getset] |
Get/set the negative overlap upperbound for the unmatched predictions (default = 0.5). More... | |
PriorBoxParameter.CodeType | code_type [getset] |
Get/set the coding method for the bounding box. More... | |
bool | encode_variance_in_target [getset] |
Get/set whether or not to encode the variance of the prior box in the loc loss target instead of in the bbox (default = false). More... | |
bool | map_object_to_agnostic [getset] |
Get/set whether or not to map all object classes to an agnostic class (default = false). This is useful when learning objectness detector. More... | |
bool | ignore_cross_boundary_bbox [getset] |
Get/set whether or not to ignore cross boundary bbox during matching (default = false). The cross boundary bbox is a bbox who is outside of the image region. More... | |
bool | bp_inside [getset] |
Get/set whether or not to only backpropagate on corners which are inside of the image region when encode type is CORNER or CORNER_SIZE (default = false). More... | |
MiningType | mining_type [getset] |
Get/set the mining type used during training (default = MAX_NEGATIVE). More... | |
NonMaximumSuppressionParameter | nms_param [getset] |
Get/set the parameters used for the non maximum suppression during hard example training. More... | |
int | sample_size [getset] |
Get/set the number of samples (default = 64). More... | |
bool | use_prior_for_nms [getset] |
Get/set whether or not to use the prior bbox for nms. More... | |
bool | use_prior_for_matching [getset] |
Get/set whether or not to use prior for matching. More... | |
bool | use_gpu [getset] |
Use the GPU version of the algorithm. More... | |
Specifies the parameters for the MultiBoxLossParameter.
Definition at line 22 of file MultiBoxLossParameter.cs.
Defines the confidence loss types.
Enumerator | |
---|---|
SOFTMAX | Specifies to use softmax. |
LOGISTIC | Specifies to use logistic. |
Definition at line 86 of file MultiBoxLossParameter.cs.
Defines the localization loss types.
Enumerator | |
---|---|
L2 | Specifies to use L2 loss. |
SMOOTH_L1 | Specifies to use smooth L1 loss. |
Definition at line 51 of file MultiBoxLossParameter.cs.
Defines the matching method used during training.
Enumerator | |
---|---|
BIPARTITE | Specifies to use Bi-Partite. |
PER_PREDICTION | Specifies to use per-prediction matching. |
Definition at line 121 of file MultiBoxLossParameter.cs.
Defines the mining type used during training.
Enumerator | |
---|---|
NONE | Use all negatives. |
MAX_NEGATIVE | Select negatives based on the score. |
HARD_EXAMPLE | Select hard examples based on Shrivastava et. al. method.
|
Definition at line 157 of file MultiBoxLossParameter.cs.
MyCaffe.param.ssd.MultiBoxLossParameter.MultiBoxLossParameter | ( | ) |
The constructor.
Definition at line 203 of file MultiBoxLossParameter.cs.
|
virtual |
Creates a new copy of this instance of the parameter.
Implements MyCaffe.param.LayerParameterBase.
Definition at line 495 of file MultiBoxLossParameter.cs.
|
static |
Convert a string into a ConfLossType.
str | Specifies the string. |
Definition at line 103 of file MultiBoxLossParameter.cs.
|
virtual |
Copy on parameter to another.
src | Specifies the parameter to copy. |
Implements MyCaffe.param.LayerParameterBase.
Definition at line 462 of file MultiBoxLossParameter.cs.
|
static |
Parses the parameter from a RawProto.
rp | Specifies the RawProto to parse. |
Definition at line 543 of file MultiBoxLossParameter.cs.
|
virtual |
Load the parameter from a binary reader.
br | Specifies the binary reader. |
bNewInstance | When true a new instance is created (the default), otherwise the existing instance is loaded from the binary reader. |
Implements MyCaffe.param.LayerParameterBase.
Definition at line 447 of file MultiBoxLossParameter.cs.
|
static |
Convert a string into a LocLossType.
str | Specifies the string. |
Definition at line 68 of file MultiBoxLossParameter.cs.
|
static |
Convert a string into a MatchType.
str | Specifies the string. |
Definition at line 138 of file MultiBoxLossParameter.cs.
|
static |
Convert a string into a MiningType.
str | Specifies the string. |
Definition at line 181 of file MultiBoxLossParameter.cs.
|
virtual |
Convert the parameter into a RawProto.
strName | Specifies the name to associate with the RawProto. |
Implements MyCaffe.basecode.BaseParameter.
Definition at line 507 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the background label id.
Definition at line 281 of file MultiBoxLossParameter.cs.
|
getset |
Get/set whether or not to only backpropagate on corners which are inside of the image region when encode type is CORNER or CORNER_SIZE (default = false).
Definition at line 375 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the coding method for the bounding box.
Definition at line 334 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the confidence loss type (default = SOFTMAX).
Definition at line 221 of file MultiBoxLossParameter.cs.
|
getset |
DEPRECIATED: Get/set whether or not to perform negative mining (default = false).
DEPRECIATED: using 'mining_type' instead.
Definition at line 304 of file MultiBoxLossParameter.cs.
|
getset |
Get/set whether or not to encode the variance of the prior box in the loc loss target instead of in the bbox (default = false).
Definition at line 344 of file MultiBoxLossParameter.cs.
|
getset |
Get/set whether or not to ignore cross boundary bbox during matching (default = false). The cross boundary bbox is a bbox who is outside of the image region.
Definition at line 365 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the localization loss type (default = SMOOTH_L1).
Definition at line 211 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the weight for the localization loss (default = 1.0).
Definition at line 231 of file MultiBoxLossParameter.cs.
|
getset |
Get/set whether or not to map all object classes to an agnostic class (default = false). This is useful when learning objectness detector.
Definition at line 354 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the matching method used during training (default = PER_PREDICTION).
Definition at line 261 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the mining type used during training (default = MAX_NEGATIVE).
Definition at line 385 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the negative overlap upperbound for the unmatched predictions (default = 0.5).
Definition at line 324 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the negative/positive ratio (default = 3.0).
Definition at line 314 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the parameters used for the non maximum suppression during hard example training.
Definition at line 395 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the number of classes to be predicted - required!
Definition at line 241 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the overlap threshold (default = 0.5).
Definition at line 271 of file MultiBoxLossParameter.cs.
|
getset |
Get/set the number of samples (default = 64).
Definition at line 405 of file MultiBoxLossParameter.cs.
|
getset |
Get/sets whether or not the bounding box is shared among different classes (default = true).
Definition at line 251 of file MultiBoxLossParameter.cs.
|
getset |
Get/set whether or not to consider the difficult ground truth (defalt = true).
Definition at line 291 of file MultiBoxLossParameter.cs.
|
getset |
Use the GPU version of the algorithm.
Definition at line 435 of file MultiBoxLossParameter.cs.
|
getset |
Get/set whether or not to use prior for matching.
Definition at line 425 of file MultiBoxLossParameter.cs.
|
getset |
Get/set whether or not to use the prior bbox for nms.
Definition at line 415 of file MultiBoxLossParameter.cs.