doc update

[jacinto-ai/caffe-jacinto.git] / src / caffe / proto / caffe.proto

syntax = "proto2";

package caffe;

// Math and storage types
enum Type {
  DOUBLE = 0;
  FLOAT = 1;
  FLOAT16 = 2;
  INT = 3;  // math not supported
  UINT = 4;  // math not supported
  BOOL = 5; //math not supported
}

enum Packing {
  NCHW = 0;
  NHWC = 1;
}

// 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];
  // New raw storage (faster and takes 1/2 of space for FP16)
  optional Type raw_data_type = 10;
  optional Type raw_diff_type = 11;
  optional bytes raw_data = 12 [packed = false];
  optional bytes raw_diff = 13 [packed = false];
  // 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];
  // Unique record index assigned by Reader
  optional uint32 record_id = 8 [default = 0];
}

// 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;
  
  // Added for keypoints information for each bbox. It is in normalized form [0-1]
  repeated float keypoint_x = 9;
  repeated float keypoint_y = 10;
}

// 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;
  // Unique record index assigned by Reader
  optional uint32 record_id = 4 [default = 0];  
}

enum DatumTypeInfo {
  DatumTypeInfo_DATUM = 0;
  DatumTypeInfo_ANNOTATED_DATUM = 1;
}

// Caffe 2 datasets support
message C2TensorProto {
  // The dimensions in the tensor.
  repeated int64 dims = 1;
  enum DataType {
    UNDEFINED = 0;
    FLOAT = 1;  // float
    INT32 = 2;  // int
    BYTE = 3;  // BYTE, when deserialized, is going to be restored as uint8.
    STRING = 4;  // string
    // Less-commonly used data types.
    BOOL = 5;  // bool
    UINT8 = 6;  // uint8_t
    INT8 = 7;  // int8_t
    UINT16 = 8;  // uint16_t
    INT16 = 9;  // int16_t
    INT64 = 10;  // int64_t
    FLOAT16 = 12;  // caffe2::__f16, caffe2::float16
    DOUBLE = 13;  // double
  }
  optional DataType data_type = 2 [default = FLOAT];
  // For float
  repeated float float_data = 3 [packed = true];
  // For int32, uint8, int8, uint16, int16, bool, and float16
  // Note about float16: in storage we will basically convert float16 byte-wise
  // to unsigned short and then store them in the int32_data field.
  repeated int32 int32_data = 4 [packed = true];
  // For bytes
  optional bytes byte_data = 5;
  // For strings
  repeated bytes string_data = 6;
  // For double
  repeated double double_data = 9 [packed = true];
  // For int64
  repeated int64 int64_data = 10 [packed = true];
  // Optionally, a name for the tensor.
  optional string name = 7;
}

message C2TensorProtos {
  repeated C2TensorProto protos = 1;
}


message FillerParameter {
  // The filler type. Can be one of the following:
  // constant, gaussian, positive_unitball, uniform, xavier,
  // msra, bilinear, or their static versions:
  // gaussianstatic, positive_unitballstatic, uniformstatic, 
  // xavierstatic, msrastatic, bilinearstatic
  // In the static version the random number generator is only
  // called once, and subsequent calls to fill blob, will write
  // the same random numbers as in the first call. This is done
  // to save time in cases where having same random numbers does
  // not make a difference.
  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;

  // Default types for all layers
  // These work only when layer-specific ones are omitted
  optional Type default_forward_type  = 11 [default = FLOAT];
  optional Type default_backward_type = 12 [default = FLOAT];
  optional Type default_forward_math  = 13 [default = FLOAT];
  optional Type default_backward_math = 14 [default = FLOAT];
  
  // Global gradient scaling coefficient K (default - no scaling)
  //
  // Scenario 1: global_grad_scale_adaptive: true
  // If positive, gradients scaled by K*L where L is L_2 norm of all gradients in a Net.
  // This helps to improve accuracy of reduced precision training.
  //
  // Scenario 2: global_grad_scale_adaptive: false
  // If positive, gradients scaled by K.
  optional float global_grad_scale = 15 [default = 1.];
  optional bool global_grad_scale_adaptive = 16 [default = false];

  // Sets the default "conv_algos_override" value for every convolution layer
  optional string default_conv_algos_override = 17 [default = "-1,-1,-1"];

  // While using multiple GPUs we have to run reduction process after every iteration.
  // For better performance we unify multiple layers in buckets.
  // This parameter sets approximate number of buckets to combine layers to.
  // Default value is good for majority of nets.
  optional int32 reduce_buckets = 18 [default = 3];

  // Sets the default "cudnn_math_override" value for every layer
  optional int32 default_cudnn_math_override = 19 [default = -1];
  
  optional bool quantize = 200 [default = false];  
  optional NetQuantizationParameter net_quantization_param = 201;  
}

// NOTE
// Update the next available ID when you add a new SolverParameter field.
//
// SolverParameter next available ID: 55 (last added: test_and_snapshot_last_epochs)
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 = 241 [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 = 242 [default = "Integral"];
  // If true, display per class result.
  optional bool show_per_class_result = 244 [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 = false];
  
  optional int32 rampup_interval = 41 [default = 0];
  optional float rampup_lr = 42 [default = 0.];
  optional float min_lr = 43 [default = 0.];
  
  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 string momentum_policy = 46 [default = "fixed"];  
  optional float max_momentum     = 47 [default = 0.99];
  optional float momentum_power   = 48 [default = 1.];
  
  // LARC - Layer-wise Adaptive Rate Control
  optional bool larc = 49 [default = false]; 
  optional string larc_policy = 50 [default = "scale"];
  optional float  larc_eta = 51    [default = 0.001];
  
  optional float  weight_decay = 12; // The weight decay.  
  optional string weight_decay_policy = 52 [default = "fixed"]; 
  optional float  weight_decay_power  = 53 [default = 0.5];
  
  // 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 = 243;

  // 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];
  // Type used for storing weights and history
  optional Type solver_data_type = 44 [default = FLOAT];
  // If true:
  // * Stores blobs in old (less efficient) BVLC-compatible format.
  // * FP16 blobs are converted to FP32 and stored in 'data' container.
  // * FP32 blobs are stored in 'data' container.
  // * FP64 blobs are stored in 'double_data' container.
  optional bool store_blobs_in_old_format = 45 [default = false];
  // If set to N>0, makes Caffe to test and snapshot last N epochs
  optional int32 test_and_snapshot_last_epochs = 54 [default = 0];
  // Ignore mismatching blobs and continue while loading weights
  optional bool ignore_shape_mismatch = 150 [default = true];
    
  // Wite additional snapshot in txt format
  optional bool snapshot_log = 151 [default = false];

  // Sparsity params
  optional int32 display_sparsity = 152 [default = 0];
  optional SparseMode sparse_mode = 153 [default = SPARSE_NONE];  
  optional float sparsity_target = 154 [default = 0.0]; //desired sparsity as a fraction
  optional float sparsity_step_factor = 155 [default = 0.01];  //sprsity step factor. 0.01 is 1%
  optional int32 sparsity_step_iter = 156 [default = 1000];  //sprsity step increment iterations
  optional int32 sparsity_start_iter = 157 [default = 0];  //sprsity start iteration  
  optional float sparsity_start_factor = 158 [default = 0.0];  //sprsity start factor
  optional float sparsity_threshold_maxratio = 159 [default = 0.2]; //default ratio of the max threshold allowed to the max weight
  optional bool sparsity_itr_increment_bfr_applying = 160 [default = true]; //true:itr increment before applying sparsity(old behaviour), false: after applying sparsity
  optional float sparsity_threshold_value_max = 161 [default = 0.2]; //threshold_value_max
}

// 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: 152 (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

  // Type returned by Forward routines of a particular layer (aka 'Ftype')
  optional Type forward_type = 145 [default = FLOAT];
  // Type returned by Backward routines of a particular layer (aka 'Btype')
  optional Type backward_type = 146 [default = FLOAT];
  // Internal math types. Works for those layers where internal math type 
  // could be different compared to Ftype or Btype. For example, so called
  // "pseudo fp32 mode" in convolution layers. For other layers has no meaning. 
  optional Type forward_math = 147 [default = FLOAT];
  optional Type backward_math = 148 [default = FLOAT];

  optional bool debug = 149 [default = false];

  // Sets the default cudnnMathType_t value for all cuDNN-based
  // computations in current lyer, if applicable. Ignored otherwise.
  // If negative or omitted, assumes implicit default and allows
  // optimizers like cudnnFindConvolution*AlgorithmEx to choose the best type.
  // If set to zero, enforces using CUDNN_DEFAULT_MATH everywhere in current lyer.
  // If set to one, enforces using CUDNN_TENSOR_OP_MATH everywhere in current lyer.
  optional int32 cudnn_math_override = 150 [default = -1];

  // 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 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 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 RecurrentParameter recurrent_param = 151;

  // NVIDIA PARAMETERS (Start with 68 because NV is 68 on an old-style phone)
  optional DetectNetGroundTruthParameter detectnet_groundtruth_param = 6801;
  optional DetectNetAugmentationParameter detectnet_augmentation_param = 6802;

  // TI PARAMETERS (Start with 84 because TI is 84 on an old-style phone)   
  optional ImageLabelDataParameter image_label_data_param = 8403;
  optional QuantizationParameter quantization_param = 8404;
}

// Message that stores parameters used to apply transformation
// to the data layer's data
message TransformationParameter {
  enum InterpolationAlgo {
    INTER_NEAREST = 0; //!< nearest neighbor interpolation
    INTER_LINEAR  = 1; //!< bilinear interpolation
    INTER_CUBIC   = 2; //!< bicubic interpolation
    INTER_AREA    = 3; //!< area-based (or super) interpolation
  }
  // When the images in a batch are of different shapes, we need to preprocess
  // them into the same fixed shape, as downstream operations in caffe require
  // images within a batch to be of the same shape.
  //
  // To transform one image of arbitrary shape into an image of fixed shape,
  // we allow specifying a sequence of "variable-sized image transforms."
  // There are three possible transforms, and it is possible for _all of them_
  // to be enabled at the same time. They are always applied in the same order:
  // (1) first random resize, (2) then random crop, (3) finally center crop.
  // The last transform must be either a random crop or a center crop.
  //
  // The three supported transforms are as follows:
  //
  // 1. Random resize. This takes two parameters, "lower" and "upper," or
  //    "L" and "U" for short. If the original image has shape (oldW, oldH),
  //    the shorter side, D = min(oldW, oldH), is calculated. Then a resize
  //    target size R is chosing uniformly from the interval [L, U], and both
  //    sides of the original image are resized by a scaling factor R/D to yield
  //    a new image with shape (R/D * oldW, R/D * oldH).
  //
  // 2. Random crop. This takes one 'crop_size' parameter. A square region is randomly
  //    chosen from the image for cropping. Works in TRAIN phase only.
  //
  // 3. Center crop. This takes one 'crop_size' parameter. A square region is chosen
  //    from the center of the image for cropping. Works in TEST phase only.
  //
  optional uint32 img_rand_resize_lower = 10 [default = 0];
  optional uint32 img_rand_resize_upper = 11 [default = 0];

  // Limits for randomly generated ratio R so that longer side
  // length would be set to shorter side length multiplied by R.
  // If applied to sguare, the shorter side is chosen randomly.
  // Pair {1,1} means "resize image to square (by shortest side)".
  // Values less than 1 are ignored.
  optional float rand_resize_ratio_lower = 12 [default = 0];
  optional float rand_resize_ratio_upper = 13 [default = 0];

  // Limits for randomly generated vertical stretch, i.e.
  // "height" *= "vertical_stretch" where
  // "vertical_stretch" = Rand(vertical_stretch_lower,vertical_stretch_upper).
  // Pair {1,1} means "do nothing".
  optional float vertical_stretch_lower = 14 [default = 1];
  optional float vertical_stretch_upper = 15 [default = 1];

  // Limits for randomly generated horizontal stretch, i.e.
  // "width" *= "horizontal_stretch" where
  // "horizontal_stretch" = Rand(horizontal_stretch_lower,horizontal_stretch_upper).
  // Pair {1,1} means "do nothing".
  optional float horizontal_stretch_lower = 16 [default = 1];
  optional float horizontal_stretch_upper = 17 [default = 1];

  // OpenCV algorithm used for downsampling
  optional InterpolationAlgo interpolation_algo_down = 18 [default = INTER_NEAREST];
  // OpenCV algorithm used for upsampling
  optional InterpolationAlgo interpolation_algo_up = 19 [default = INTER_CUBIC];
  // No upscale by default no matter what resize parameters are chosen
  optional bool allow_upscale = 20 [default = false];

  // If followed by CuDNN, set to NHWC for better performance
  optional Packing forward_packing = 21 [default = NCHW];

  // 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 = 211 [default = 0];
  optional uint32 crop_w = 212 [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];
  // Run the transform (synchronously) on the GPU
  // False if omitted when Forward Type is float/double.
  // True otherwise (float16 doesn't work well on CPU).
  optional bool use_gpu_transform = 8 [default = false];
  // If non-negative, the seed with which the transformer's
  // random number generator would be initialized -- useful for reproducible results.
  // Otherwise, (and by default) initialize using a seed derived from the system clock.
  optional int64 random_seed = 9 [default = -1];

  optional bool display = 22 [default = false];
  optional int32 num_labels = 23 [default = 0];
  
  // Resize policy
  optional ResizeParameter resize_param = 208;
  // Noise policy
  optional NoiseParameter noise_param = 209;
  // Distortion policy
  optional DistortionParameter distort_param = 213;
  // Expand policy
  optional ExpansionParameter expand_param = 214;
  // Constraint for emitting the annotation after transformation.
  optional EmitConstraint emit_constraint = 210;  
}
// Message that stores parameters used to create gridbox ground truth
message DetectNetGroundTruthParameter {
  // stride of gridbox with respect to image size
  optional uint32 stride = 1 [default = 4];
  // coverage region scale with respect to bounding box size
  optional float scale_cvg = 2 [default = 0.5];
  enum GridboxType {
    GRIDBOX_MAX = 0;
    GRIDBOX_MIN = 1;
  }
  // determines coverage region's maximum and minimum dimensions
  optional GridboxType gridbox_type = 3 [default = GRIDBOX_MAX];
  // if gridbox_type is equal to GRIDBOX_MAX, the maximum size a given coverage
  //  region may take.
  optional uint32 max_cvg_len = 4 [default = 50];
  // if gridbox_type is equal to GRIDBOX_MIN, the minimum size a given coverage
  //  region may take.
  optional uint32 min_cvg_len = 5 [default = 50];
  enum CoverageType {
    RECTANGULAR = 0;
  }
  // shape of the coverage geometry.
  optional CoverageType coverage_type = 7 [default = RECTANGULAR];
  // Size that incoming images are cropped / scaled to during training / test
  //  time. The network will only see images of this size.
  optional uint32 image_size_x = 8 [default = 1248];
  optional uint32 image_size_y = 9 [default = 384];
  // coverage proportional to the size of the gridbox region overlapping the
  // normalize object coverage by the size of the object
  optional bool obj_norm = 11 [default = false];
  // crop incoming bboxes such that their bounds remain inside the gridbox.
  optional bool crop_bboxes = 12 [default = true];
  // Integer and target index of classes to be included:
  message ClassMapping {
    required uint32 src = 1;
    required uint32 dst = 2;
  }
  repeated ClassMapping object_class = 13;
}

// Message that stores parameters used to apply image and label augmentations
// specific to NVDataLayer's online augmentation module
message DetectNetAugmentationParameter {
  // probability that a random crop of the training image will be taken at test
  //  time. If image dimensions are less than input size and random cropping is
  //  off, then the image will be scaled deterministically to the input size.
  optional float crop_prob = 1 [default = 1.0];
  // number of pixels to shift the image by. If cropping is enabled, this number
  //  is added to the range of possible crop offset values.
  optional uint32 shift_x = 2 [default = 0];
  optional uint32 shift_y = 3 [default = 0];
  // probability that the training image will be scaled at test time. 0 turns
  //  scale augmentation off.
  optional float scale_prob = 4 [default = 0.33];
  // minimum and maximum scaling factor. 1.0 implies no scaling.
  optional float scale_min = 5 [default = 0.7];
  optional float scale_max = 6 [default = 1.0];
  // probability that image will be flipped across the Y axis. 0 turns flip
  //  augmentation off.
  optional float flip_prob = 7 [default = 0.33];
  // maximum angle in degrees (in both directions) image may be rotated. 0 turns
  //  rotation augmentation off.
  optional float rotation_prob = 8 [default = 0.33];
  optional float max_rotate_degree = 9 [default = 1.0];
  // maximum rotation of hue in degrees (in both directions). 0 turns hue
  //  augmentation off.
  optional float hue_rotation_prob = 10 [default = 0.33];
  optional float hue_rotation = 11 [default = 15];
  // maximum desaturation parameter. 1.0 may convert RGB to luminance. 0 turns
  //  desaturation augmentation off.
  optional float desaturation_prob = 12 [default = 0.33];
  optional float desaturation_max = 13 [default = 0.5];
  
  // Resize policy
  optional ResizeParameter resize_param = 208;
  // Noise policy
  optional NoiseParameter noise_param = 209;
  // Distortion policy
  optional DistortionParameter distort_param = 213;
  // Expand policy
  optional ExpansionParameter expand_param = 214;
  // Constraint for emitting the annotation after transformation.
  optional EmitConstraint emit_constraint = 210;  
}

// 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 layer.
  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;
  }
  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;
  
  // label weights
  repeated float label_weights = 4;
}

// 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 [default = false];
  // 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];
  optional FillerParameter scale_filler = 5;
  optional FillerParameter bias_filler = 6;
  optional bool scale_bias = 7 [default = false];
  enum Engine {
    DEFAULT = 0;
    CAFFE = 1;
    CUDNN = 2;
  }
  optional Engine engine = 15 [default = DEFAULT];
}

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];
  enum CuDNNConvolutionAlgorithmSeeker {
    GET = 0;
    FINDEX = 1;
  }
  //Specifies which cudnn routine should be used to find the best convolution algorithm
  optional CuDNNConvolutionAlgorithmSeeker cudnn_convolution_algo_seeker = 19 [default = FINDEX];

  // If set to a non-negative value, enforces using the algo by the index provided.
  // It has priority over CuDNNConvolutionAlgorithmSeeker and essentially disables seeking.
  // The index should correspond to the ordinal in structures cudnnConvolutionFwdAlgo_t,
  // cudnnConvolutionBwdDataAlgo_t and cudnnConvolutionBwdFilterAlgo_t.
  // For example, conv_algos_override set to "7,5,7" means using CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD_NONFUSED,
  // CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD_NONFUSED and CUDNN_CONVOLUTION_BWD_FILTER_ALGO_WINOGRAD_NONFUSED
  // correspondingly.
  optional string conv_algos_override = 20 [default = "-1,-1,-1"];
}

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];
  // Deprecated
  optional uint32 prefetch = 10 [default = 0];
  // Number of Data Transformer threads per GPU. If 0 Caffe optimizes it automatically.
  // Takes effect only when used together with 'parser_threads' setting.
  // Ignored and always set to 1 for test nets.
  optional uint32 threads  = 11 [default = 0];
  // Number of Parser threads per GPU. If 0 Caffe optimizes it automatically.
  // Takes effect only when used together with 'threads' setting.
  // Ignored and always set to 1 for test nets.
  optional uint32 parser_threads  = 12 [default = 0];
  // Cache observations while reading
  optional bool cache = 13 [default = false];
  // Shuffle observations while reading for better accuracy. Ignored if 'cache' is false.
  optional bool shuffle = 14 [default = false];
}

// 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;
  optional int32 num_keypoint = 7 [default = 0];
}

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;
  // Total number of keypoints predicted along with each bounding box.
  optional uint32 num_keypoint = 13 [default = 0];
}

message DropoutParameter {
  optional float dropout_ratio = 1 [default = 0.5]; // dropout ratio
  enum Engine {
    DEFAULT = 0;
    CAFFE = 1;
    CUDNN = 2;
  }
  optional Engine engine = 2 [default = DEFAULT];
  optional int64 random_seed = 3 [default = -1];
}

// 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.];
  optional float lambda = 2 [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 = ""];
  // Resize while reading to adjust images by short side.
  // Has priority over new_height and new_width.
  optional uint32 short_side = 13 [default = 0];
  // Number of Data Transformer threads per GPU.
  optional uint32 threads  = 14 [default = 4];
  // Whether or not ImageLayer should cache all files in RAM. Cache might fail if it doesn't fit.
  optional bool cache = 15 [default = false];
}

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];
  optional bool ignore_difficult_gt = 24 [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 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];
  
  // FB pooling parameters
  // Use floor((height + 2 * padding - kernel) / stride) + 1
  // instead of ceil((height + 2 * padding - kernel) / stride) + 1
  optional bool torch_pooling = 40 [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];
  optional uint32 num_keypoint = 14 [default = 0];
}

message PythonParameter {
  optional string module = 1;
  optional string layer = 2;
  // This value is set to the attribute `param_str` of the `PythonLayer` object
  // in Python before calling the `setup()` method. This could be a number,
  // string, dictionary in Python dict format, JSON, etc. You may parse this
  // string in `setup` method and use it in `forward` and `backward`.
  optional string param_str = 3 [default = ''];
  // Whether this PythonLayer is shared among worker solvers during data parallelism.
  // If true, each worker solver sequentially run forward from this layer.
  // This value should be set true if you are using it as a data layer.
  optional bool share_in_parallel = 4 [default = false];
}

// Message that stores parameters used by RecurrentLayer
message RecurrentParameter {
  // The dimension of the output (and usually hidden state) representation --
  // must be explicitly set to non-zero.
  optional uint32 num_output = 1 [default = 0];

  optional FillerParameter weight_filler = 2; // The filler for the weight
  optional FillerParameter bias_filler = 3; // The filler for the bias

  // Whether to enable displaying debug_info in the unrolled recurrent net.
  optional bool debug_info = 4 [default = false];

  // Whether to add as additional inputs (bottoms) the initial hidden state
  // blobs, and add as additional outputs (tops) the final timestep hidden state
  // blobs.  The number of additional bottom/top blobs required depends on the
  // recurrent architecture -- e.g., 1 for RNNs, 2 for LSTMs.
  optional bool expose_hidden = 5 [default = false];
}

// Message that stores parameters used by ReductionLayer
message ReductionParameter {
  enum ReductionOp {
    SUM = 1;
    ASUM = 2;
    SUMSQ = 3;
    MEAN = 4;
  }

  optional ReductionOp operation = 1 [default = SUM]; // reduction operation

  // The first axis to reduce to a scalar -- may be negative to index from the
  // end (e.g., -1 for the last axis).
  // (Currently, only reduction along ALL "tail" axes is supported; reduction
  // of axis M through N, where N < num_axes - 1, is unsupported.)
  // Suppose we have an n-axis bottom Blob with shape:
  //     (d0, d1, d2, ..., d(m-1), dm, d(m+1), ..., d(n-1)).
  // If axis == m, the output Blob will have shape
  //     (d0, d1, d2, ..., d(m-1)),
  // and the ReductionOp operation is performed (d0 * d1 * d2 * ... * d(m-1))
  // times, each including (dm * d(m+1) * ... * d(n-1)) individual data.
  // If axis == 0 (the default), the output Blob always has the empty shape
  // (count 1), performing reduction across the entire input --
  // often useful for creating new loss functions.
  optional int32 axis = 2 [default = 0];

  optional float coeff = 3 [default = 1.0]; // coefficient for output
}

// Message that stores parameters used by ReLULayer
message ReLUParameter {
  // Allow non-zero slope for negative inputs to speed up optimization
  // Described in:
  // Maas, A. L., Hannun, A. Y., & Ng, A. Y. (2013). Rectifier nonlinearities
  // improve neural network acoustic models. In ICML Workshop on Deep Learning
  // for Audio, Speech, and Language Processing.
  optional float negative_slope = 1 [default = 0];
  enum Engine {
    DEFAULT = 0;
    CAFFE = 1;
    CUDNN = 2;
  }
  optional Engine engine = 2 [default = DEFAULT];
}

message ReshapeParameter {
  // Specify the output dimensions. If some of the dimensions are set to 0,
  // the corresponding dimension from the bottom layer is used (unchanged).
  // Exactly one dimension may be set to -1, in which case its value is
  // inferred from the count of the bottom blob and the remaining dimensions.
  // For example, suppose we want to reshape a 2D blob "input" with shape 2 x 8:
  //
  //   layer {
  //     type: "Reshape" bottom: "input" top: "output"
  //     reshape_param { ... }
  //   }
  //
  // If "input" is 2D with shape 2 x 8, then the following reshape_param
  // specifications are all equivalent, producing a 3D blob "output" with shape
  // 2 x 2 x 4:
  //
  //   reshape_param { shape { dim:  2  dim: 2  dim:  4 } }
  //   reshape_param { shape { dim:  0  dim: 2  dim:  4 } }
  //   reshape_param { shape { dim:  0  dim: 2  dim: -1 } }
  //   reshape_param { shape { dim:  0  dim:-1  dim:  4 } }
  //
  optional BlobShape shape = 1;

  // axis and num_axes control the portion of the bottom blob's shape that are
  // replaced by (included in) the reshape. By default (axis == 0 and
  // num_axes == -1), the entire bottom blob shape is included in the reshape,
  // and hence the shape field must specify the entire output shape.
  //
  // axis may be non-zero to retain some portion of the beginning of the input
  // shape (and may be negative to index from the end; e.g., -1 to begin the
  // reshape after the last axis, including nothing in the reshape,
  // -2 to include only the last axis, etc.).
  //
  // For example, suppose "input" is a 2D blob with shape 2 x 8.
  // Then the following ReshapeLayer specifications are all equivalent,
  // producing a blob "output" with shape 2 x 2 x 4:
  //
  //   reshape_param { shape { dim: 2  dim: 2  dim: 4 } }
  //   reshape_param { shape { dim: 2  dim: 4 } axis:  1 }
  //   reshape_param { shape { dim: 2  dim: 4 } axis: -3 }
  //
  // num_axes specifies the extent of the reshape.
  // If num_axes >= 0 (and axis >= 0), the reshape will be performed only on
  // input axes in the range [axis, axis+num_axes].
  // num_axes may also be -1, the default, to include all remaining axes
  // (starting from axis).
  //
  // For example, suppose "input" is a 2D blob with shape 2 x 8.
  // Then the following ReshapeLayer specifications are equivalent,
  // producing a blob "output" with shape 1 x 2 x 8.
  //
  //   reshape_param { shape { dim:  1  dim: 2  dim:  8 } }
  //   reshape_param { shape { dim:  1  dim: 2  }  num_axes: 1 }
  //   reshape_param { shape { dim:  1  }  num_axes: 0 }
  //
  // On the other hand, these would produce output blob shape 2 x 1 x 8:
  //
  //   reshape_param { shape { dim: 2  dim: 1  dim: 8  }  }
  //   reshape_param { shape { dim: 1 }  axis: 1  num_axes: 0 }
  //
  optional int32 axis = 2 [default = 0];
  optional int32 num_axes = 3 [default = -1];
}

message ScaleParameter {
  // The first axis of bottom[0] (the first input Blob) along which to apply
  // bottom[1] (the second input Blob).  May be negative to index from the end
  // (e.g., -1 for the last axis).
  //
  // For example, if bottom[0] is 4D with shape 100x3x40x60, the output
  // top[0] will have the same shape, and bottom[1] may have any of the
  // following shapes (for the given value of axis):
  //    (axis == 0 == -4) 100; 100x3; 100x3x40; 100x3x40x60
  //    (axis == 1 == -3)          3;     3x40;     3x40x60
  //    (axis == 2 == -2)                   40;       40x60
  //    (axis == 3 == -1)                                60
  // Furthermore, bottom[1] may have the empty shape (regardless of the value of
  // "axis") -- a scalar multiplier.
  optional int32 axis = 1 [default = 1];

  // (num_axes is ignored unless just one bottom is given and the scale is
  // a learned parameter of the layer.  Otherwise, num_axes is determined by the
  // number of axes by the second bottom.)
  // The number of axes of the input (bottom[0]) covered by the scale
  // parameter, or -1 to cover all axes of bottom[0] starting from `axis`.
  // Set num_axes := 0, to multiply with a zero-axis Blob: a scalar.
  optional int32 num_axes = 2 [default = 1];

  // (filler is ignored unless just one bottom is given and the scale is
  // a learned parameter of the layer.)
  // The initialization for the learned scale parameter.
  // Default is the unit (1) initialization, resulting in the ScaleLayer
  // initially performing the identity operation.
  optional FillerParameter filler = 3;

  // Whether to also learn a bias (equivalent to a ScaleLayer+BiasLayer, but
  // may be more efficient).  Initialized with bias_filler (defaults to 0).
  optional bool bias_term = 4 [default = false];
  optional FillerParameter bias_filler = 5;
}

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 parameters are shared across channels.
  optional bool channel_shared = 2 [default = false];
}

// Specifies how to deal with connections whose weights are recovered/initialized as zeros. 
// Used for fine-tuning
enum SparseMode {
  SPARSE_NONE = 0;               //keep connected
  SPARSE_UPDATE = 1;             //disconnect all zero-weight connections during back propagation udpate
} 

// Message for layers with reduced word with arithmetic
enum QuantizationParameter_Precision {
  QuantizationParameter_Precision_FLOAT = 0;    
  QuantizationParameter_Precision_DYNAMIC_FIXED_POINT = 1;
}
enum QuantizationParameter_Rounding {
  QuantizationParameter_Rounding_NEAREST = 0;
  QuantizationParameter_Rounding_STOCHASTIC = 1;
}

// Quantization params for a layer
message QuantizationParameter {  
  optional QuantizationParameter_Precision precision = 1 [default = QuantizationParameter_Precision_DYNAMIC_FIXED_POINT];  
  optional QuantizationParameter_Rounding rounding_scheme = 2 [default = QuantizationParameter_Rounding_NEAREST];
  optional bool power2_scale_weights = 3 [default = false];
  optional bool power2_scale_activations = 4 [default = false];  
          
  // Dynamic fixed point params
  message QParams {
    optional bool quantize = 1 [default = false];  
    optional uint32 bitwidth = 2 [default = 8];    
    optional int32 fracbits = 3 [default = 0];
    optional bool unsigned_data = 4 [ default = false];
    optional bool unsigned_quant = 5 [ default = false];    
    optional float scale_target = 6 [default = 1.0];
    optional float scale_applied = 7 [default = 1.0];    
    optional int32 shiftbits = 8 [default = 0];      
    optional float offset = 9 [default = 0];    
    optional float min = 10 [default = 0];
    optional float max = 11 [default = 0];    
  }
  repeated QParams qparam_in = 5;
  repeated QParams qparam_w = 6;
  repeated QParams qparam_out = 7;  
  
  optional int32 quantized_infer_count = 8 [ default = 0];  
}

// Quantization params for the net  
message NetQuantizationParameter {
  optional bool quantize_weights = 1 [default = true];
  optional bool quantize_activations = 2 [default = true];  
          
  //8-bits doesn't seem to be working for classification if previous frame's range is used.  
  optional uint32 bitwidth_weights = 3 [default = 12];
  optional uint32 bitwidth_activations = 4 [default = 8];
  optional uint32 bitwidth_bias = 5 [default = 16];
        
  //indicates whether quantized range is power of 2 or not. 
  //if input is quantized with power2 range, the output quantized value can be obtained by shifting with fracbits
  optional bool power2_scale_weights = 6 [default = false];
  optional bool power2_scale_activations = 7 [default = false];
      
  //apply offset to make quantized range unsigned and optimal.
  optional bool apply_offset_weights = 8 [default = true];
  optional bool apply_offset_activations = 9 [default = false];  
  
  optional QuantizationParameter_Precision precision = 10 [default = QuantizationParameter_Precision_DYNAMIC_FIXED_POINT];  
  optional QuantizationParameter_Rounding rounding_scheme = 11 [default = QuantizationParameter_Rounding_NEAREST];  
    
  optional int32 display_quantization = 12 [default = 2000];
  optional bool update_quantization_param = 13 [default = true];
  
  //frame/iter at which quantization is introduced    
  optional int32 quantization_start = 14 [default = 1];  
  optional float range_update_factor = 15 [default = 0.01];
  optional float range_expansion_factor = 16 [default = 1.25]; 
  
  repeated string ignored_layer_names = 17;
}

message ImageSliceInfo {
  repeated uint32 dim = 1;
  repeated uint32 stride = 2;
  repeated uint32 offset = 3;
}


message ImageLabelDataParameter {
  enum DB {
    LEVELDB = 0;
    LMDB = 1;
  }
  // Specify the data source.
  optional string image_list_path = 1;
  optional string label_list_path = 2;
  // Specify the batch size.
  optional uint32 batch_size = 3 [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 = 4 [default = 0];
  // Whether or not ImageLayer should shuffle the list of files at every epoch.
  optional bool shuffle = 5 [default = false];

  // Whether random sample image scales
  optional float scale_prob = 6 [default = 0];
  optional float scale_min = 7  [default = 1];
  optional float scale_max = 8  [default = 1];

  optional string data_dir = 9 [default = ""];

  optional ImageSliceInfo label_slice = 10;

  optional string image_dir = 11 [default = ""];
  optional string label_dir = 12 [default = ""];

  enum Padding {
    ZERO = 0;
    REFLECT = 1;
  }

  optional Padding padding = 13 [default = REFLECT];
  
  optional string image_dir2 = 14 [default = ""];  
  optional string image_list_path2 = 15;  
  
  optional uint32 threads = 16 [default = 0];
  optional uint32 size_min = 17 [default = 0];  
  optional uint32 size_max = 18 [default = 0]; 
  
  optional bool check_image_files = 19 [default = false];
  
  optional DB backend = 20 [default = LEVELDB];  
}