You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@singa.apache.org by wa...@apache.org on 2016/10/18 13:07:30 UTC
[1/3] incubator-singa git commit: SINGA-253 Net converter for caffe
model
Repository: incubator-singa
Updated Branches:
refs/heads/master c967169bc -> 32b4368f0
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/src/proto/caffe.proto
----------------------------------------------------------------------
diff --git a/src/proto/caffe.proto b/src/proto/caffe.proto
new file mode 100644
index 0000000..82fabdb
--- /dev/null
+++ b/src/proto/caffe.proto
@@ -0,0 +1,1402 @@
+// Version of caffe.proto:
+// commits on Jun 29, 2016 in caffe github reporsitory
+
+syntax = "proto2";
+
+package caffe;
+
+// Specifies the shape (dimensions) of a Blob.
+message BlobShape {
+ repeated int64 dim = 1 [packed = true];
+}
+
+message BlobProto {
+ optional BlobShape shape = 7;
+ repeated float data = 5 [packed = true];
+ repeated float diff = 6 [packed = true];
+ repeated double double_data = 8 [packed = true];
+ repeated double double_diff = 9 [packed = true];
+
+ // 4D dimensions -- deprecated. Use "shape" instead.
+ optional int32 num = 1 [default = 0];
+ optional int32 channels = 2 [default = 0];
+ optional int32 height = 3 [default = 0];
+ optional int32 width = 4 [default = 0];
+}
+
+// The BlobProtoVector is simply a way to pass multiple blobproto instances
+// around.
+message BlobProtoVector {
+ repeated BlobProto blobs = 1;
+}
+
+message Datum {
+ optional int32 channels = 1;
+ optional int32 height = 2;
+ optional int32 width = 3;
+ // the actual image data, in bytes
+ optional bytes data = 4;
+ optional int32 label = 5;
+ // Optionally, the datum could also hold float data.
+ repeated float float_data = 6;
+ // If true data contains an encoded image that need to be decoded
+ optional bool encoded = 7 [default = false];
+}
+
+message FillerParameter {
+ // The filler type.
+ optional string type = 1 [default = 'constant'];
+ optional float value = 2 [default = 0]; // the value in constant filler
+ optional float min = 3 [default = 0]; // the min value in uniform filler
+ optional float max = 4 [default = 1]; // the max value in uniform filler
+ optional float mean = 5 [default = 0]; // the mean value in Gaussian filler
+ optional float std = 6 [default = 1]; // the std value in Gaussian filler
+ // The expected number of non-zero output weights for a given input in
+ // Gaussian filler -- the default -1 means don't perform sparsification.
+ optional int32 sparse = 7 [default = -1];
+ // Normalize the filler variance by fan_in, fan_out, or their average.
+ // Applies to 'xavier' and 'msra' fillers.
+ enum VarianceNorm {
+ FAN_IN = 0;
+ FAN_OUT = 1;
+ AVERAGE = 2;
+ }
+ optional VarianceNorm variance_norm = 8 [default = FAN_IN];
+}
+
+message NetParameter {
+ optional string name = 1; // consider giving the network a name
+ // DEPRECATED. See InputParameter. The input blobs to the network.
+ repeated string input = 3;
+ // DEPRECATED. See InputParameter. The shape of the input blobs.
+ repeated BlobShape input_shape = 8;
+
+ // 4D input dimensions -- deprecated. Use "input_shape" instead.
+ // If specified, for each input blob there should be four
+ // values specifying the num, channels, height and width of the input blob.
+ // Thus, there should be a total of (4 * #input) numbers.
+ repeated int32 input_dim = 4;
+
+ // Whether the network will force every layer to carry out backward operation.
+ // If set False, then whether to carry out backward is determined
+ // automatically according to the net structure and learning rates.
+ optional bool force_backward = 5 [default = false];
+ // The current "state" of the network, including the phase, level, and stage.
+ // Some layers may be included/excluded depending on this state and the states
+ // specified in the layers' include and exclude fields.
+ optional NetState state = 6;
+
+ // Print debugging information about results while running Net::Forward,
+ // Net::Backward, and Net::Update.
+ optional bool debug_info = 7 [default = false];
+
+ // The layers that make up the net. Each of their configurations, including
+ // connectivity and behavior, is specified as a LayerParameter.
+ repeated LayerParameter layer = 100; // ID 100 so layers are printed last.
+
+ // DEPRECATED: use 'layer' instead.
+ repeated V1LayerParameter layers = 2;
+}
+
+// NOTE
+// Update the next available ID when you add a new SolverParameter field.
+//
+// SolverParameter next available ID: 41 (last added: type)
+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;
+
+ // The number of iterations for each test net.
+ repeated int32 test_iter = 3;
+
+ // The number of iterations between two testing phases.
+ optional int32 test_interval = 4 [default = 0];
+ optional bool test_compute_loss = 19 [default = false];
+ // If true, run an initial test pass before the first iteration,
+ // ensuring memory availability and printing the starting value of the loss.
+ optional bool test_initialization = 32 [default = true];
+ optional float base_lr = 5; // The base learning rate
+ // the number of iterations between displaying info. If display = 0, no info
+ // will be displayed.
+ optional int32 display = 6;
+ // Display the loss averaged over the last average_loss iterations
+ optional int32 average_loss = 33 [default = 1];
+ optional int32 max_iter = 7; // the maximum number of iterations
+ // accumulate gradients over `iter_size` x `batch_size` instances
+ optional int32 iter_size = 36 [default = 1];
+
+ // The learning rate decay policy. The currently implemented learning rate
+ // policies are as follows:
+ // - fixed: always return base_lr.
+ // - step: return base_lr * gamma ^ (floor(iter / step))
+ // - exp: return base_lr * gamma ^ iter
+ // - inv: return base_lr * (1 + gamma * iter) ^ (- power)
+ // - multistep: similar to step but it allows non uniform steps defined by
+ // stepvalue
+ // - poly: the effective learning rate follows a polynomial decay, to be
+ // zero by the max_iter. return base_lr (1 - iter/max_iter) ^ (power)
+ // - sigmoid: the effective learning rate follows a sigmod decay
+ // return base_lr ( 1/(1 + exp(-gamma * (iter - stepsize))))
+ //
+ // where base_lr, max_iter, gamma, step, stepvalue and power are defined
+ // in the solver parameter protocol buffer, and iter is the current iteration.
+ optional string lr_policy = 8;
+ optional float gamma = 9; // The parameter to compute the learning rate.
+ optional float power = 10; // The parameter to compute the learning rate.
+ optional float momentum = 11; // The momentum value.
+ optional float weight_decay = 12; // The weight decay.
+ // regularization types supported: L1 and L2
+ // controlled by weight_decay
+ optional string regularization_type = 29 [default = "L2"];
+ // the stepsize for learning rate policy "step"
+ optional int32 stepsize = 13;
+ // the stepsize for learning rate policy "multistep"
+ repeated int32 stepvalue = 34;
+
+ // Set clip_gradients to >= 0 to clip parameter gradients to that L2 norm,
+ // whenever their actual L2 norm is larger.
+ optional float clip_gradients = 35 [default = -1];
+
+ optional int32 snapshot = 14 [default = 0]; // The snapshot interval
+ optional string snapshot_prefix = 15; // The prefix for the snapshot.
+ // whether to snapshot diff in the results or not. Snapshotting diff will help
+ // debugging but the final protocol buffer size will be much larger.
+ optional bool snapshot_diff = 16 [default = false];
+ enum SnapshotFormat {
+ HDF5 = 0;
+ BINARYPROTO = 1;
+ }
+ optional SnapshotFormat snapshot_format = 37 [default = BINARYPROTO];
+ // the mode solver will use: 0 for CPU and 1 for GPU. Use GPU in default.
+ enum SolverMode {
+ CPU = 0;
+ GPU = 1;
+ }
+ optional SolverMode solver_mode = 17 [default = GPU];
+ // the device_id will that be used in GPU mode. Use device_id = 0 in default.
+ optional int32 device_id = 18 [default = 0];
+ // If non-negative, the seed with which the Solver will initialize the Caffe
+ // random number generator -- useful for reproducible results. Otherwise,
+ // (and by default) initialize using a seed derived from the system clock.
+ optional int64 random_seed = 20 [default = -1];
+
+ // type of the solver
+ optional string type = 40 [default = "SGD"];
+
+ // numerical stability for RMSProp, AdaGrad and AdaDelta and Adam
+ optional float delta = 31 [default = 1e-8];
+ // parameters for the Adam solver
+ optional float momentum2 = 39 [default = 0.999];
+
+ // RMSProp decay value
+ // MeanSquare(t) = rms_decay*MeanSquare(t-1) + (1-rms_decay)*SquareGradient(t)
+ optional float rms_decay = 38 [default = 0.99];
+
+ // If true, print information about the state of the net that may help with
+ // debugging learning problems.
+ optional bool debug_info = 23 [default = false];
+
+ // If false, don't save a snapshot after training finishes.
+ optional bool snapshot_after_train = 28 [default = true];
+
+ // DEPRECATED: old solver enum types, use string instead
+ enum SolverType {
+ SGD = 0;
+ NESTEROV = 1;
+ ADAGRAD = 2;
+ RMSPROP = 3;
+ ADADELTA = 4;
+ ADAM = 5;
+ }
+ // DEPRECATED: use type instead of solver_type
+ optional SolverType solver_type = 30 [default = SGD];
+}
+
+// A message that stores the solver snapshots
+message SolverState {
+ optional int32 iter = 1; // The current iteration
+ optional string learned_net = 2; // The file that stores the learned net.
+ repeated BlobProto history = 3; // The history for sgd solvers
+ optional int32 current_step = 4 [default = 0]; // The current step for learning rate
+}
+
+enum Phase {
+ TRAIN = 0;
+ TEST = 1;
+}
+
+message NetState {
+ optional Phase phase = 1 [default = TEST];
+ optional int32 level = 2 [default = 0];
+ repeated string stage = 3;
+}
+
+message NetStateRule {
+ // Set phase to require the NetState have a particular phase (TRAIN or TEST)
+ // to meet this rule.
+ optional Phase phase = 1;
+
+ // Set the minimum and/or maximum levels in which the layer should be used.
+ // Leave undefined to meet the rule regardless of level.
+ optional int32 min_level = 2;
+ optional int32 max_level = 3;
+
+ // Customizable sets of stages to include or exclude.
+ // The net must have ALL of the specified stages and NONE of the specified
+ // "not_stage"s to meet the rule.
+ // (Use multiple NetStateRules to specify conjunctions of stages.)
+ repeated string stage = 4;
+ repeated string not_stage = 5;
+}
+
+// Specifies training parameters (multipliers on global learning constants,
+// and the name and other settings used for weight sharing).
+message ParamSpec {
+ // The names of the parameter blobs -- useful for sharing parameters among
+ // layers, but never required otherwise. To share a parameter between two
+ // layers, give it a (non-empty) name.
+ optional string name = 1;
+
+ // Whether to require shared weights to have the same shape, or just the same
+ // count -- defaults to STRICT if unspecified.
+ optional DimCheckMode share_mode = 2;
+ enum DimCheckMode {
+ // STRICT (default) requires that num, channels, height, width each match.
+ STRICT = 0;
+ // PERMISSIVE requires only the count (num*channels*height*width) to match.
+ PERMISSIVE = 1;
+ }
+
+ // The multiplier on the global learning rate for this parameter.
+ optional float lr_mult = 3 [default = 1.0];
+
+ // The multiplier on the global weight decay for this parameter.
+ optional float decay_mult = 4 [default = 1.0];
+}
+
+// NOTE
+// Update the next available ID when you add a new LayerParameter field.
+//
+// LayerParameter next available layer-specific ID: 147 (last added: recurrent_param)
+message LayerParameter {
+ optional string name = 1; // the layer name
+ optional string type = 2; // the layer type
+ repeated string bottom = 3; // the name of each bottom blob
+ repeated string top = 4; // the name of each top blob
+
+ // The train / test phase for computation.
+ optional Phase phase = 10;
+
+ // The amount of weight to assign each top blob in the objective.
+ // Each layer assigns a default value, usually of either 0 or 1,
+ // to each top blob.
+ repeated float loss_weight = 5;
+
+ // Specifies training parameters (multipliers on global learning constants,
+ // and the name and other settings used for weight sharing).
+ repeated ParamSpec param = 6;
+
+ // The blobs containing the numeric parameters of the layer.
+ repeated BlobProto blobs = 7;
+
+ // Specifies whether to backpropagate to each bottom. If unspecified,
+ // Caffe will automatically infer whether each input needs backpropagation
+ // to compute parameter gradients. If set to true for some inputs,
+ // backpropagation to those inputs is forced; if set false for some inputs,
+ // backpropagation to those inputs is skipped.
+ //
+ // The size must be either 0 or equal to the number of bottoms.
+ repeated bool propagate_down = 11;
+
+ // Rules controlling whether and when a layer is included in the network,
+ // based on the current NetState. You may specify a non-zero number of rules
+ // to include OR exclude, but not both. If no include or exclude rules are
+ // specified, the layer is always included. If the current NetState meets
+ // ANY (i.e., one or more) of the specified rules, the layer is
+ // included/excluded.
+ repeated NetStateRule include = 8;
+ repeated NetStateRule exclude = 9;
+
+ // Parameters for data pre-processing.
+ optional TransformationParameter transform_param = 100;
+
+ // Parameters shared by loss layers.
+ optional LossParameter loss_param = 101;
+
+ // Layer type-specific parameters.
+ //
+ // Note: certain layers may have more than one computational engine
+ // for their implementation. These layers include an Engine type and
+ // engine parameter for selecting the implementation.
+ // The default for the engine is set by the ENGINE switch at compile-time.
+ optional AccuracyParameter accuracy_param = 102;
+ optional 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 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 MVNParameter mvn_param = 120;
+ optional ParameterParameter parameter_param = 145;
+ optional PoolingParameter pooling_param = 121;
+ optional PowerParameter power_param = 122;
+ optional PReLUParameter prelu_param = 131;
+ optional PythonParameter python_param = 130;
+ optional RecurrentParameter recurrent_param = 146;
+ optional ReductionParameter reduction_param = 136;
+ optional ReLUParameter relu_param = 123;
+ optional ReshapeParameter reshape_param = 133;
+ optional ScaleParameter scale_param = 142;
+ optional SigmoidParameter sigmoid_param = 124;
+ optional SoftmaxParameter softmax_param = 125;
+ optional SPPParameter spp_param = 132;
+ optional SliceParameter slice_param = 126;
+ optional TanHParameter tanh_param = 127;
+ optional ThresholdParameter threshold_param = 128;
+ optional TileParameter tile_param = 138;
+ optional WindowDataParameter window_data_param = 129;
+}
+
+// Message that stores parameters used to apply transformation
+// to the data layer's data
+message TransformationParameter {
+ // For data pre-processing, we can do simple scaling and subtracting the
+ // data mean, if provided. Note that the mean subtraction is always carried
+ // out before scaling.
+ optional float scale = 1 [default = 1];
+ // Specify if we want to randomly mirror data.
+ optional bool mirror = 2 [default = false];
+ // Specify if we would like to randomly crop an image.
+ optional uint32 crop_size = 3 [default = 0];
+ // 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];
+}
+
+// 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;
+}
+
+// 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 ArgMaxParameter {
+ // If true produce pairs (argmax, maxval)
+ optional bool out_max_val = 1 [default = false];
+ optional uint32 top_k = 2 [default = 1];
+ // The axis along which to maximise -- may be negative to index from the
+ // end (e.g., -1 for the last axis).
+ // By default ArgMaxLayer maximizes over the flattened trailing dimensions
+ // for each index of the first / num dimension.
+ optional int32 axis = 3;
+}
+
+message ConcatParameter {
+ // The axis along which to concatenate -- may be negative to index from the
+ // end (e.g., -1 for the last axis). Other axes must have the
+ // same dimension for all the bottom blobs.
+ // By default, ConcatLayer concatenates blobs along the "channels" axis (1).
+ optional int32 axis = 2 [default = 1];
+
+ // DEPRECATED: alias for "axis" -- does not support negative indexing.
+ optional uint32 concat_dim = 1 [default = 1];
+}
+
+message BatchNormParameter {
+ // If false, accumulate global mean/variance values via a moving average. If
+ // true, use those accumulated values instead of computing mean/variance
+ // across the batch.
+ optional bool use_global_stats = 1;
+ // How much does the moving average decay each iteration?
+ optional float moving_average_fraction = 2 [default = .999];
+ // Small value to add to the variance estimate so that we don't divide by
+ // zero.
+ optional float eps = 3 [default = 1e-5];
+}
+
+message BiasParameter {
+ // The first axis of bottom[0] (the first input Blob) along which to apply
+ // bottom[1] (the second input Blob). May be negative to index from the end
+ // (e.g., -1 for the last axis).
+ //
+ // For example, if bottom[0] is 4D with shape 100x3x40x60, the output
+ // top[0] will have the same shape, and bottom[1] may have any of the
+ // following shapes (for the given value of axis):
+ // (axis == 0 == -4) 100; 100x3; 100x3x40; 100x3x40x60
+ // (axis == 1 == -3) 3; 3x40; 3x40x60
+ // (axis == 2 == -2) 40; 40x60
+ // (axis == 3 == -1) 60
+ // Furthermore, bottom[1] may have the empty shape (regardless of the value of
+ // "axis") -- a scalar bias.
+ optional int32 axis = 1 [default = 1];
+
+ // (num_axes is ignored unless just one bottom is given and the bias is
+ // a learned parameter of the layer. Otherwise, num_axes is determined by the
+ // number of axes by the second bottom.)
+ // The number of axes of the input (bottom[0]) covered by the bias
+ // parameter, or -1 to cover all axes of bottom[0] starting from `axis`.
+ // Set num_axes := 0, to add a zero-axis Blob: a scalar.
+ optional int32 num_axes = 2 [default = 1];
+
+ // (filler is ignored unless just one bottom is given and the bias is
+ // a learned parameter of the layer.)
+ // The initialization for the learned bias parameter.
+ // Default is the zero (0) initialization, resulting in the BiasLayer
+ // initially performing the identity operation.
+ optional FillerParameter filler = 3;
+}
+
+message ContrastiveLossParameter {
+ // margin for dissimilar pair
+ optional float margin = 1 [default = 1.0];
+ // The first implementation of this cost did not exactly match the cost of
+ // Hadsell et al 2006 -- using (margin - d^2) instead of (margin - d)^2.
+ // legacy_version = false (the default) uses (margin - d)^2 as proposed in the
+ // Hadsell paper. New models should probably use this version.
+ // legacy_version = true uses (margin - d^2). This is kept to support /
+ // reproduce existing models and results
+ optional bool legacy_version = 2 [default = false];
+}
+
+message ConvolutionParameter {
+ optional uint32 num_output = 1; // The number of outputs for the layer
+ optional bool bias_term = 2 [default = true]; // whether to have bias terms
+
+ // Pad, kernel size, and stride are all given as a single value for equal
+ // dimensions in all spatial dimensions, or once per spatial dimension.
+ repeated uint32 pad = 3; // The padding size; defaults to 0
+ repeated uint32 kernel_size = 4; // The kernel size
+ repeated uint32 stride = 6; // The stride; defaults to 1
+ // Factor used to dilate the kernel, (implicitly) zero-filling the resulting
+ // holes. (Kernel dilation is sometimes referred to by its use in the
+ // algorithme � trous from Holschneider et al. 1987.)
+ repeated uint32 dilation = 18; // The dilation; defaults to 1
+
+ // For 2D convolution only, the *_h and *_w versions may also be used to
+ // specify both spatial dimensions.
+ optional uint32 pad_h = 9 [default = 0]; // The padding height (2D only)
+ optional uint32 pad_w = 10 [default = 0]; // The padding width (2D only)
+ optional uint32 kernel_h = 11; // The kernel height (2D only)
+ optional uint32 kernel_w = 12; // The kernel width (2D only)
+ optional uint32 stride_h = 13; // The stride height (2D only)
+ optional uint32 stride_w = 14; // The stride width (2D only)
+
+ optional uint32 group = 5 [default = 1]; // The group size for group conv
+
+ optional FillerParameter weight_filler = 7; // The filler for the weight
+ optional FillerParameter bias_filler = 8; // The filler for the bias
+ enum Engine {
+ DEFAULT = 0;
+ CAFFE = 1;
+ CUDNN = 2;
+ }
+ optional Engine engine = 15 [default = DEFAULT];
+
+ // The axis to interpret as "channels" when performing convolution.
+ // Preceding dimensions are treated as independent inputs;
+ // succeeding dimensions are treated as "spatial".
+ // With (N, C, H, W) inputs, and axis == 1 (the default), we perform
+ // N independent 2D convolutions, sliding C-channel (or (C/g)-channels, for
+ // groups g>1) filters across the spatial axes (H, W) of the input.
+ // With (N, C, D, H, W) inputs, and axis == 1, we perform
+ // N independent 3D convolutions, sliding (C/g)-channels
+ // filters across the spatial axes (D, H, W) of the input.
+ optional int32 axis = 16 [default = 1];
+
+ // Whether to force use of the general ND convolution, even if a specific
+ // implementation for blobs of the appropriate number of spatial dimensions
+ // is available. (Currently, there is only a 2D-specific convolution
+ // implementation; for input blobs with num_axes != 2, this option is
+ // ignored and the ND implementation will be used.)
+ optional bool force_nd_im2col = 17 [default = false];
+}
+
+message CropParameter {
+ // To crop, elements of the first bottom are selected to fit the dimensions
+ // of the second, reference bottom. The crop is configured by
+ // - the crop `axis` to pick the dimensions for cropping
+ // - the crop `offset` to set the shift for all/each dimension
+ // to align the cropped bottom with the reference bottom.
+ // All dimensions up to but excluding `axis` are preserved, while
+ // the dimensions including and trailing `axis` are cropped.
+ // If only one `offset` is set, then all dimensions are offset by this amount.
+ // Otherwise, the number of offsets must equal the number of cropped axes to
+ // shift the crop in each dimension accordingly.
+ // Note: standard dimensions are N,C,H,W so the default is a spatial crop,
+ // and `axis` may be negative to index from the end (e.g., -1 for the last
+ // axis).
+ optional int32 axis = 1 [default = 2];
+ repeated uint32 offset = 2;
+}
+
+message DataParameter {
+ enum DB {
+ LEVELDB = 0;
+ LMDB = 1;
+ }
+ // Specify the data source.
+ optional string source = 1;
+ // Specify the batch size.
+ optional uint32 batch_size = 4;
+ // The rand_skip variable is for the data layer to skip a few data points
+ // to avoid all asynchronous sgd clients to start at the same point. The skip
+ // point would be set as rand_skip * rand(0,1). Note that rand_skip should not
+ // be larger than the number of keys in the database.
+ // DEPRECATED. Each solver accesses a different subset of the database.
+ optional uint32 rand_skip = 7 [default = 0];
+ optional DB backend = 8 [default = LEVELDB];
+ // DEPRECATED. See TransformationParameter. For data pre-processing, we can do
+ // simple scaling and subtracting the data mean, if provided. Note that the
+ // mean subtraction is always carried out before scaling.
+ optional float scale = 2 [default = 1];
+ optional string mean_file = 3;
+ // DEPRECATED. See TransformationParameter. Specify if we would like to randomly
+ // crop an image.
+ optional uint32 crop_size = 5 [default = 0];
+ // DEPRECATED. See TransformationParameter. Specify if we want to randomly mirror
+ // data.
+ optional bool mirror = 6 [default = false];
+ // Force the encoded image to have 3 color channels
+ optional bool force_encoded_color = 9 [default = false];
+ // Prefetch queue (Number of batches to prefetch to host memory, increase if
+ // data access bandwidth varies).
+ optional uint32 prefetch = 10 [default = 4];
+}
+
+message DropoutParameter {
+ optional float dropout_ratio = 1 [default = 0.5]; // dropout ratio
+}
+
+// DummyDataLayer fills any number of arbitrarily shaped blobs with random
+// (or constant) data generated by "Fillers" (see "message FillerParameter").
+message DummyDataParameter {
+ // This layer produces N >= 1 top blobs. DummyDataParameter must specify 1 or N
+ // shape fields, and 0, 1 or N data_fillers.
+ //
+ // If 0 data_fillers are specified, ConstantFiller with a value of 0 is used.
+ // If 1 data_filler is specified, it is applied to all top blobs. If N are
+ // specified, the ith is applied to the ith top blob.
+ repeated FillerParameter data_filler = 1;
+ repeated BlobShape shape = 6;
+
+ // 4D dimensions -- deprecated. Use "shape" instead.
+ repeated uint32 num = 2;
+ repeated uint32 channels = 3;
+ repeated uint32 height = 4;
+ repeated uint32 width = 5;
+}
+
+message EltwiseParameter {
+ enum EltwiseOp {
+ PROD = 0;
+ SUM = 1;
+ MAX = 2;
+ }
+ optional EltwiseOp operation = 1 [default = SUM]; // element-wise operation
+ repeated float coeff = 2; // blob-wise coefficient for SUM operation
+
+ // Whether to use an asymptotically slower (for >2 inputs) but stabler method
+ // of computing the gradient for the PROD operation. (No effect for SUM op.)
+ optional bool stable_prod_grad = 3 [default = true];
+}
+
+// Message that stores parameters used by ELULayer
+message ELUParameter {
+ // Described in:
+ // Clevert, D.-A., Unterthiner, T., & Hochreiter, S. (2015). Fast and Accurate
+ // Deep Network Learning by Exponential Linear Units (ELUs). arXiv
+ optional float alpha = 1 [default = 1];
+}
+
+// Message that stores parameters used by EmbedLayer
+message EmbedParameter {
+ optional uint32 num_output = 1; // The number of outputs for the layer
+ // The input is given as integers to be interpreted as one-hot
+ // vector indices with dimension num_input. Hence num_input should be
+ // 1 greater than the maximum possible input value.
+ optional uint32 input_dim = 2;
+
+ optional bool bias_term = 3 [default = true]; // Whether to use a bias term
+ optional FillerParameter weight_filler = 4; // The filler for the weight
+ optional FillerParameter bias_filler = 5; // The filler for the bias
+
+}
+
+// Message that stores parameters used by ExpLayer
+message ExpParameter {
+ // ExpLayer computes outputs y = base ^ (shift + scale * x), for base > 0.
+ // Or if base is set to the default (-1), base is set to e,
+ // so y = exp(shift + scale * x).
+ optional float base = 1 [default = -1.0];
+ optional float scale = 2 [default = 1.0];
+ optional float shift = 3 [default = 0.0];
+}
+
+/// Message that stores parameters used by FlattenLayer
+message FlattenParameter {
+ // The first axis to flatten: all preceding axes are retained in the output.
+ // May be negative to index from the end (e.g., -1 for the last axis).
+ optional int32 axis = 1 [default = 1];
+
+ // The last axis to flatten: all following axes are retained in the output.
+ // May be negative to index from the end (e.g., the default -1 for the last
+ // axis).
+ optional int32 end_axis = 2 [default = -1];
+}
+
+// Message that stores parameters used by HDF5DataLayer
+message HDF5DataParameter {
+ // Specify the data source.
+ optional string source = 1;
+ // Specify the batch size.
+ optional uint32 batch_size = 2;
+
+ // Specify whether to shuffle the data.
+ // If shuffle == true, the ordering of the HDF5 files is shuffled,
+ // and the ordering of data within any given HDF5 file is shuffled,
+ // but data between different files are not interleaved; all of a file's
+ // data are output (in a random order) before moving onto another file.
+ optional bool shuffle = 3 [default = false];
+}
+
+message HDF5OutputParameter {
+ optional string file_name = 1;
+}
+
+message HingeLossParameter {
+ enum Norm {
+ L1 = 1;
+ L2 = 2;
+ }
+ // Specify the Norm to use L1 or L2
+ optional Norm norm = 1 [default = L1];
+}
+
+message ImageDataParameter {
+ // Specify the data source.
+ optional string source = 1;
+ // Specify the batch size.
+ optional uint32 batch_size = 4 [default = 1];
+ // The rand_skip variable is for the data layer to skip a few data points
+ // to avoid all asynchronous sgd clients to start at the same point. The skip
+ // point would be set as rand_skip * rand(0,1). Note that rand_skip should not
+ // be larger than the number of keys in the database.
+ optional uint32 rand_skip = 7 [default = 0];
+ // Whether or not ImageLayer should shuffle the list of files at every epoch.
+ optional bool shuffle = 8 [default = false];
+ // It will also resize images if new_height or new_width are not zero.
+ optional uint32 new_height = 9 [default = 0];
+ optional uint32 new_width = 10 [default = 0];
+ // Specify if the images are color or gray
+ optional bool is_color = 11 [default = true];
+ // DEPRECATED. See TransformationParameter. For data pre-processing, we can do
+ // simple scaling and subtracting the data mean, if provided. Note that the
+ // mean subtraction is always carried out before scaling.
+ optional float scale = 2 [default = 1];
+ optional string mean_file = 3;
+ // DEPRECATED. See TransformationParameter. Specify if we would like to randomly
+ // crop an image.
+ optional uint32 crop_size = 5 [default = 0];
+ // DEPRECATED. See TransformationParameter. Specify if we want to randomly mirror
+ // data.
+ optional bool mirror = 6 [default = false];
+ optional string root_folder = 12 [default = ""];
+}
+
+message InfogainLossParameter {
+ // Specify the infogain matrix source.
+ optional string source = 1;
+}
+
+message InnerProductParameter {
+ optional uint32 num_output = 1; // The number of outputs for the layer
+ optional bool bias_term = 2 [default = true]; // whether to have bias terms
+ optional FillerParameter weight_filler = 3; // The filler for the weight
+ optional FillerParameter bias_filler = 4; // The filler for the bias
+
+ // The first axis to be lumped into a single inner product computation;
+ // all preceding axes are retained in the output.
+ // May be negative to index from the end (e.g., -1 for the last axis).
+ optional int32 axis = 5 [default = 1];
+ // Specify whether to transpose the weight matrix or not.
+ // If transpose == true, any operations will be performed on the transpose
+ // of the weight matrix. The weight matrix itself is not going to be transposed
+ // but rather the transfer flag of operations will be toggled accordingly.
+ optional bool transpose = 6 [default = false];
+}
+
+message InputParameter {
+ // This layer produces N >= 1 top blob(s) to be assigned manually.
+ // Define N shapes to set a shape for each top.
+ // Define 1 shape to set the same shape for every top.
+ // Define no shape to defer to reshaping manually.
+ repeated BlobShape shape = 1;
+}
+
+// Message that stores parameters used by LogLayer
+message LogParameter {
+ // LogLayer computes outputs y = log_base(shift + scale * x), for base > 0.
+ // Or if base is set to the default (-1), base is set to e,
+ // so y = ln(shift + scale * x) = log_e(shift + scale * x)
+ optional float base = 1 [default = -1.0];
+ optional float scale = 2 [default = 1.0];
+ optional float shift = 3 [default = 0.0];
+}
+
+// Message that stores parameters used by LRNLayer
+message LRNParameter {
+ optional uint32 local_size = 1 [default = 5];
+ optional float alpha = 2 [default = 1.];
+ optional float beta = 3 [default = 0.75];
+ enum NormRegion {
+ ACROSS_CHANNELS = 0;
+ WITHIN_CHANNEL = 1;
+ }
+ optional NormRegion norm_region = 4 [default = ACROSS_CHANNELS];
+ optional float k = 5 [default = 1.];
+ enum Engine {
+ DEFAULT = 0;
+ CAFFE = 1;
+ CUDNN = 2;
+ }
+ optional Engine engine = 6 [default = DEFAULT];
+}
+
+message MemoryDataParameter {
+ optional uint32 batch_size = 1;
+ optional uint32 channels = 2;
+ optional uint32 height = 3;
+ optional uint32 width = 4;
+}
+
+message 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 ParameterParameter {
+ optional BlobShape shape = 1;
+}
+
+message PoolingParameter {
+ enum PoolMethod {
+ MAX = 0;
+ AVE = 1;
+ STOCHASTIC = 2;
+ }
+ optional PoolMethod pool = 1 [default = MAX]; // The pooling method
+ // Pad, kernel size, and stride are all given as a single value for equal
+ // dimensions in height and width or as Y, X pairs.
+ optional uint32 pad = 4 [default = 0]; // The padding size (equal in Y, X)
+ optional uint32 pad_h = 9 [default = 0]; // The padding height
+ optional uint32 pad_w = 10 [default = 0]; // The padding width
+ optional uint32 kernel_size = 2; // The kernel size (square)
+ optional uint32 kernel_h = 5; // The kernel height
+ optional uint32 kernel_w = 6; // The kernel width
+ optional uint32 stride = 3 [default = 1]; // The stride (equal in Y, X)
+ optional uint32 stride_h = 7; // The stride height
+ optional uint32 stride_w = 8; // The stride width
+ enum Engine {
+ DEFAULT = 0;
+ CAFFE = 1;
+ CUDNN = 2;
+ }
+ optional Engine engine = 11 [default = DEFAULT];
+ // If global_pooling then it will pool over the size of the bottom by doing
+ // kernel_h = bottom->height and kernel_w = bottom->width
+ optional bool global_pooling = 12 [default = false];
+}
+
+message PowerParameter {
+ // PowerLayer computes outputs y = (shift + scale * x) ^ power.
+ optional float power = 1 [default = 1.0];
+ optional float scale = 2 [default = 1.0];
+ optional float shift = 3 [default = 0.0];
+}
+
+message 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 WindowDataParameter {
+ // Specify the data source.
+ optional string source = 1;
+ // For data pre-processing, we can do simple scaling and subtracting the
+ // data mean, if provided. Note that the mean subtraction is always carried
+ // out before scaling.
+ optional float scale = 2 [default = 1];
+ optional string mean_file = 3;
+ // Specify the batch size.
+ optional uint32 batch_size = 4;
+ // Specify if we would like to randomly crop an image.
+ optional uint32 crop_size = 5 [default = 0];
+ // Specify if we want to randomly mirror data.
+ optional bool mirror = 6 [default = false];
+ // Foreground (object) overlap threshold
+ optional float fg_threshold = 7 [default = 0.5];
+ // Background (non-object) overlap threshold
+ optional float bg_threshold = 8 [default = 0.5];
+ // Fraction of batch that should be foreground objects
+ optional float fg_fraction = 9 [default = 0.25];
+ // Amount of contextual padding to add around a window
+ // (used only by the window_data_layer)
+ optional uint32 context_pad = 10 [default = 0];
+ // Mode for cropping out a detection window
+ // warp: cropped window is warped to a fixed size and aspect ratio
+ // square: the tightest square around the window is cropped
+ optional string crop_mode = 11 [default = "warp"];
+ // cache_images: will load all images in memory for faster access
+ optional bool cache_images = 12 [default = false];
+ // append root_folder to locate images
+ optional string root_folder = 13 [default = ""];
+}
+
+message SPPParameter {
+ enum PoolMethod {
+ MAX = 0;
+ AVE = 1;
+ STOCHASTIC = 2;
+ }
+ optional uint32 pyramid_height = 1;
+ optional PoolMethod pool = 2 [default = MAX]; // The pooling method
+ enum Engine {
+ DEFAULT = 0;
+ CAFFE = 1;
+ CUDNN = 2;
+ }
+ optional Engine engine = 6 [default = DEFAULT];
+}
+
+// DEPRECATED: use LayerParameter.
+message V1LayerParameter {
+ repeated string bottom = 2;
+ repeated string top = 3;
+ optional string name = 4;
+ repeated NetStateRule include = 32;
+ repeated NetStateRule exclude = 33;
+ enum LayerType {
+ NONE = 0;
+ ABSVAL = 35;
+ ACCURACY = 1;
+ ARGMAX = 30;
+ BNLL = 2;
+ CONCAT = 3;
+ CONTRASTIVE_LOSS = 37;
+ CONVOLUTION = 4;
+ DATA = 5;
+ DECONVOLUTION = 39;
+ DROPOUT = 6;
+ DUMMY_DATA = 32;
+ EUCLIDEAN_LOSS = 7;
+ ELTWISE = 25;
+ EXP = 38;
+ FLATTEN = 8;
+ HDF5_DATA = 9;
+ HDF5_OUTPUT = 10;
+ HINGE_LOSS = 28;
+ IM2COL = 11;
+ IMAGE_DATA = 12;
+ INFOGAIN_LOSS = 13;
+ INNER_PRODUCT = 14;
+ LRN = 15;
+ MEMORY_DATA = 29;
+ MULTINOMIAL_LOGISTIC_LOSS = 16;
+ MVN = 34;
+ POOLING = 17;
+ POWER = 26;
+ RELU = 18;
+ SIGMOID = 19;
+ SIGMOID_CROSS_ENTROPY_LOSS = 27;
+ SILENCE = 36;
+ SOFTMAX = 20;
+ SOFTMAX_LOSS = 21;
+ SPLIT = 22;
+ SLICE = 33;
+ TANH = 23;
+ WINDOW_DATA = 24;
+ THRESHOLD = 31;
+ }
+ optional LayerType type = 5;
+ repeated BlobProto blobs = 6;
+ repeated string param = 1001;
+ repeated DimCheckMode blob_share_mode = 1002;
+ enum DimCheckMode {
+ STRICT = 0;
+ PERMISSIVE = 1;
+ }
+ repeated float blobs_lr = 7;
+ repeated float weight_decay = 8;
+ repeated float loss_weight = 35;
+ optional AccuracyParameter accuracy_param = 27;
+ optional ArgMaxParameter argmax_param = 23;
+ optional ConcatParameter concat_param = 9;
+ optional ContrastiveLossParameter contrastive_loss_param = 40;
+ optional ConvolutionParameter convolution_param = 10;
+ optional DataParameter data_param = 11;
+ optional DropoutParameter dropout_param = 12;
+ optional DummyDataParameter dummy_data_param = 26;
+ optional EltwiseParameter eltwise_param = 24;
+ optional ExpParameter exp_param = 41;
+ optional HDF5DataParameter hdf5_data_param = 13;
+ optional HDF5OutputParameter hdf5_output_param = 14;
+ optional HingeLossParameter hinge_loss_param = 29;
+ optional ImageDataParameter image_data_param = 15;
+ optional InfogainLossParameter infogain_loss_param = 16;
+ optional InnerProductParameter inner_product_param = 17;
+ optional LRNParameter lrn_param = 18;
+ optional MemoryDataParameter memory_data_param = 22;
+ optional MVNParameter mvn_param = 34;
+ optional PoolingParameter pooling_param = 19;
+ optional PowerParameter power_param = 21;
+ optional ReLUParameter relu_param = 30;
+ optional SigmoidParameter sigmoid_param = 38;
+ optional SoftmaxParameter softmax_param = 39;
+ optional SliceParameter slice_param = 31;
+ optional TanHParameter tanh_param = 37;
+ optional ThresholdParameter threshold_param = 25;
+ optional WindowDataParameter window_data_param = 20;
+ optional TransformationParameter transform_param = 36;
+ optional LossParameter loss_param = 42;
+ optional V0LayerParameter layer = 1;
+}
+
+// DEPRECATED: V0LayerParameter is the old way of specifying layer parameters
+// in Caffe. We keep this message type around for legacy support.
+message V0LayerParameter {
+ optional string name = 1; // the layer name
+ optional string type = 2; // the string to specify the layer type
+
+ // Parameters to specify layers with inner products.
+ optional uint32 num_output = 3; // The number of outputs for the layer
+ optional bool biasterm = 4 [default = true]; // whether to have bias terms
+ optional FillerParameter weight_filler = 5; // The filler for the weight
+ optional FillerParameter bias_filler = 6; // The filler for the bias
+
+ optional uint32 pad = 7 [default = 0]; // The padding size
+ optional uint32 kernelsize = 8; // The kernel size
+ optional uint32 group = 9 [default = 1]; // The group size for group conv
+ optional uint32 stride = 10 [default = 1]; // The stride
+ enum PoolMethod {
+ MAX = 0;
+ AVE = 1;
+ STOCHASTIC = 2;
+ }
+ optional PoolMethod pool = 11 [default = MAX]; // The pooling method
+ optional float dropout_ratio = 12 [default = 0.5]; // dropout ratio
+
+ optional uint32 local_size = 13 [default = 5]; // for local response norm
+ optional float alpha = 14 [default = 1.]; // for local response norm
+ optional float beta = 15 [default = 0.75]; // for local response norm
+ optional float k = 22 [default = 1.];
+
+ // For data layers, specify the data source
+ optional string source = 16;
+ // For data pre-processing, we can do simple scaling and subtracting the
+ // data mean, if provided. Note that the mean subtraction is always carried
+ // out before scaling.
+ optional float scale = 17 [default = 1];
+ optional string meanfile = 18;
+ // For data layers, specify the batch size.
+ optional uint32 batchsize = 19;
+ // For data layers, specify if we would like to randomly crop an image.
+ optional uint32 cropsize = 20 [default = 0];
+ // For data layers, specify if we want to randomly mirror data.
+ optional bool mirror = 21 [default = false];
+
+ // The blobs containing the numeric parameters of the layer
+ repeated BlobProto blobs = 50;
+ // The ratio that is multiplied on the global learning rate. If you want to
+ // set the learning ratio for one blob, you need to set it for all blobs.
+ repeated float blobs_lr = 51;
+ // The weight decay that is multiplied on the global weight decay.
+ repeated float weight_decay = 52;
+
+ // The rand_skip variable is for the data layer to skip a few data points
+ // to avoid all asynchronous sgd clients to start at the same point. The skip
+ // point would be set as rand_skip * rand(0,1). Note that rand_skip should not
+ // be larger than the number of keys in the database.
+ optional uint32 rand_skip = 53 [default = 0];
+
+ // Fields related to detection (det_*)
+ // foreground (object) overlap threshold
+ optional float det_fg_threshold = 54 [default = 0.5];
+ // background (non-object) overlap threshold
+ optional float det_bg_threshold = 55 [default = 0.5];
+ // Fraction of batch that should be foreground objects
+ optional float det_fg_fraction = 56 [default = 0.25];
+
+ // optional bool OBSOLETE_can_clobber = 57 [default = true];
+
+ // Amount of contextual padding to add around a window
+ // (used only by the window_data_layer)
+ optional uint32 det_context_pad = 58 [default = 0];
+
+ // Mode for cropping out a detection window
+ // warp: cropped window is warped to a fixed size and aspect ratio
+ // square: the tightest square around the window is cropped
+ optional string det_crop_mode = 59 [default = "warp"];
+
+ // For ReshapeLayer, one needs to specify the new dimensions.
+ optional int32 new_num = 60 [default = 0];
+ optional int32 new_channels = 61 [default = 0];
+ optional int32 new_height = 62 [default = 0];
+ optional int32 new_width = 63 [default = 0];
+
+ // Whether or not ImageLayer should shuffle the list of files at every epoch.
+ // It will also resize images if new_height or new_width are not zero.
+ optional bool shuffle_images = 64 [default = false];
+
+ // For ConcatLayer, one needs to specify the dimension for concatenation, and
+ // the other dimensions must be the same for all the bottom blobs.
+ // By default it will concatenate blobs along the channels dimension.
+ optional uint32 concat_dim = 65 [default = 1];
+
+ optional HDF5OutputParameter hdf5_output_param = 1001;
+}
+
+message PReLUParameter {
+ // Parametric ReLU described in K. He et al, Delving Deep into Rectifiers:
+ // Surpassing Human-Level Performance on ImageNet Classification, 2015.
+
+ // Initial value of a_i. Default is a_i=0.25 for all i.
+ optional FillerParameter filler = 1;
+ // Whether or not slope paramters are shared across channels.
+ optional bool channel_shared = 2 [default = false];
+}
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/src/proto/model.proto
----------------------------------------------------------------------
diff --git a/src/proto/model.proto b/src/proto/model.proto
index 545b556..641efb2 100644
--- a/src/proto/model.proto
+++ b/src/proto/model.proto
@@ -217,7 +217,8 @@ message LayerConf {
optional HingeLossConf hinge_loss_conf = 114;
// optional ImageDataConf image_data_conf = 115;
optional InfogainLossConf infogain_loss_conf = 116;
- optional InnerProductConf inner_product_conf = 117;
+ // use dense_conf to replace this inner_product_conf
+ // optional InnerProductConf inner_product_conf = 117;
optional LogConf log_conf = 134;
optional LRNConf lrn_conf = 118;
// optional MemoryDataConf memory_data_conf = 119;
@@ -239,7 +240,7 @@ message LayerConf {
//optional WindowDataConf window_data_conf = 129;
// Used in SINGA
- optional DenseConf dense_conf = 201;
+ optional DenseConf dense_conf = 117;
optional MetricConf metric_conf = 200;
optional BatchNormConf batchnorm_conf = 202;
optional SplitConf split_conf = 203;
[2/3] incubator-singa git commit: SINGA-253 Net converter for caffe
model
Posted by wa...@apache.org.
SINGA-253 Net converter for caffe model
Convert caffe model into singa model.
It is a very basic implementation. Now it can convert feed forward net and supports pysinga.
Implementation method:
1. read proto file of caffe model using caffe.proto, and serialize it to string.
2. parse the string use singa model.proto and get the layer config.
3. setup each layer and add it to a feed forward net.
Update and example:
1. put conversion funcions into layer.py and converter.py;
2. add 'caffe' option in cifar example which converts alexnet from
caffe to singa.
Project: http://git-wip-us.apache.org/repos/asf/incubator-singa/repo
Commit: http://git-wip-us.apache.org/repos/asf/incubator-singa/commit/77405ec1
Tree: http://git-wip-us.apache.org/repos/asf/incubator-singa/tree/77405ec1
Diff: http://git-wip-us.apache.org/repos/asf/incubator-singa/diff/77405ec1
Branch: refs/heads/master
Commit: 77405ec19565ed82feedb933cf562f61c44522c4
Parents: c967169
Author: XiangruiCAI <ca...@gmail.com>
Authored: Thu Sep 29 22:22:00 2016 +0800
Committer: Xiangrui <ca...@gmail.com>
Committed: Tue Oct 18 11:38:26 2016 +0800
----------------------------------------------------------------------
examples/cifar10/README.md | 4 +-
examples/cifar10/caffe/__init__.py | 0
examples/cifar10/caffe/caffe_net.py | 42 +
.../cifar10/caffe/cifar10_full_solver.prototxt | 30 +
.../caffe/cifar10_full_train_test.prototxt | 223 +++
.../cifar10/caffe/cifar10_quick_solver.prototxt | 29 +
.../caffe/cifar10_quick_train_test.prototxt | 225 +++
examples/cifar10/train.py | 25 +-
python/singa/converter.py | 155 ++
python/singa/layer.py | 127 +-
src/proto/caffe.proto | 1402 ++++++++++++++++++
src/proto/model.proto | 5 +-
12 files changed, 2250 insertions(+), 17 deletions(-)
----------------------------------------------------------------------
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/examples/cifar10/README.md
----------------------------------------------------------------------
diff --git a/examples/cifar10/README.md b/examples/cifar10/README.md
index 0cf069f..65df5e6 100644
--- a/examples/cifar10/README.md
+++ b/examples/cifar10/README.md
@@ -10,6 +10,7 @@ the best validation accuracy (without data augmentation) we achieved was about 8
2. [VGGNet](http://torch.ch/blog/2015/07/30/cifar.html), the best validation accuracy (without data augmentation) we achieved was about 89%.
3. [ResNet](https://github.com/facebook/fb.resnet.torch), the best validation accuracy (without data augmentation) we achieved was about 83%.
+4. [Alexnet from Caffe](https://github.com/BVLC/caffe/tree/master/examples/cifar10), SINGA is able to convert model from Caffe seamlessly.
## Instructions
@@ -40,7 +41,8 @@ version of the Cifar-10 dataset in 'cifar-10-batches-py' folder.
python train.py vgg cifar-10-batches-py
- To train other models, please replace 'vgg' to 'alexnet' or 'resnet'. By default
+ To train other models, please replace 'vgg' to 'alexnet', 'resnet' or 'caffe',
+ where 'caffe' refers to the alexnet model converted from Caffe. By default
the training would run on a CudaGPU device, to run it on CppCPU, add an additional
argument
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/examples/cifar10/caffe/__init__.py
----------------------------------------------------------------------
diff --git a/examples/cifar10/caffe/__init__.py b/examples/cifar10/caffe/__init__.py
new file mode 100644
index 0000000..e69de29
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/examples/cifar10/caffe/caffe_net.py
----------------------------------------------------------------------
diff --git a/examples/cifar10/caffe/caffe_net.py b/examples/cifar10/caffe/caffe_net.py
new file mode 100644
index 0000000..543e3a5
--- /dev/null
+++ b/examples/cifar10/caffe/caffe_net.py
@@ -0,0 +1,42 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements. See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership. The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =============================================================================
+
+import os
+from singa import converter
+
+
+def create_net(use_cpu):
+ if use_cpu:
+ layer.engine = 'singacpp'
+
+ #net_proto = os.path.abspath('./caffe/cifar10_full_train_test.prototxt')
+ #solver_proto = os.path.abspath('./caffe/cifar10_full_solver.prototxt')
+ net_proto = os.path.abspath('./caffe/cifar10_quick_train_test.prototxt')
+ solver_proto = os.path.abspath('./caffe/cifar10_quick_solver.prototxt')
+ input_sample_shape = [3, 32, 32, ]
+
+ cvt = converter.CaffeConverter(net_proto, solver_proto, input_sample_shape)
+ net = cvt.create_net()
+ for (p, specs) in zip(net.param_values(), net.param_specs()):
+ filler = specs.filler
+ if filler.type == 'gaussian':
+ p.gaussian(filler.mean, filler.std)
+ else:
+ p.set_value(0)
+ print specs.name, filler.type, p.l1()
+
+ return net
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/examples/cifar10/caffe/cifar10_full_solver.prototxt
----------------------------------------------------------------------
diff --git a/examples/cifar10/caffe/cifar10_full_solver.prototxt b/examples/cifar10/caffe/cifar10_full_solver.prototxt
new file mode 100644
index 0000000..1e708d8
--- /dev/null
+++ b/examples/cifar10/caffe/cifar10_full_solver.prototxt
@@ -0,0 +1,30 @@
+# From caffe repository
+# Commits on Sep 2, 2015
+
+# reduce learning rate after 120 epochs (60000 iters) by factor 0f 10
+# then another factor of 10 after 10 more epochs (5000 iters)
+
+# The train/test net protocol buffer definition
+net: "examples/cifar10/cifar10_full_train_test.prototxt"
+# test_iter specifies how many forward passes the test should carry out.
+# In the case of CIFAR10, we have test batch size 100 and 100 test iterations,
+# covering the full 10,000 testing images.
+test_iter: 100
+# Carry out testing every 1000 training iterations.
+test_interval: 1000
+# The base learning rate, momentum and the weight decay of the network.
+base_lr: 0.001
+momentum: 0.9
+weight_decay: 0.004
+# The learning rate policy
+lr_policy: "fixed"
+# Display every 200 iterations
+display: 200
+# The maximum number of iterations
+max_iter: 60000
+# snapshot intermediate results
+snapshot: 10000
+snapshot_format: HDF5
+snapshot_prefix: "examples/cifar10/cifar10_full"
+# solver mode: CPU or GPU
+solver_mode: GPU
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/examples/cifar10/caffe/cifar10_full_train_test.prototxt
----------------------------------------------------------------------
diff --git a/examples/cifar10/caffe/cifar10_full_train_test.prototxt b/examples/cifar10/caffe/cifar10_full_train_test.prototxt
new file mode 100644
index 0000000..40b39ec
--- /dev/null
+++ b/examples/cifar10/caffe/cifar10_full_train_test.prototxt
@@ -0,0 +1,223 @@
+# From caffe repository
+# Commits on Feb 6, 2015
+
+name: "CIFAR10_full"
+layer {
+ name: "cifar"
+ type: "Data"
+ top: "data"
+ top: "label"
+ include {
+ phase: TRAIN
+ }
+ transform_param {
+ mean_file: "examples/cifar10/mean.binaryproto"
+ }
+ data_param {
+ source: "examples/cifar10/cifar10_train_lmdb"
+ batch_size: 100
+ backend: LMDB
+ }
+}
+layer {
+ name: "cifar"
+ type: "Data"
+ top: "data"
+ top: "label"
+ include {
+ phase: TEST
+ }
+ transform_param {
+ mean_file: "examples/cifar10/mean.binaryproto"
+ }
+ data_param {
+ source: "examples/cifar10/cifar10_test_lmdb"
+ batch_size: 100
+ backend: LMDB
+ }
+}
+layer {
+ name: "conv1"
+ type: "Convolution"
+ bottom: "data"
+ top: "conv1"
+ param {
+ lr_mult: 1
+ }
+ param {
+ lr_mult: 2
+ }
+ convolution_param {
+ num_output: 32
+ pad: 2
+ kernel_size: 5
+ stride: 1
+ weight_filler {
+ type: "gaussian"
+ std: 0.0001
+ }
+ bias_filler {
+ type: "constant"
+ }
+ }
+}
+layer {
+ name: "pool1"
+ type: "Pooling"
+ bottom: "conv1"
+ top: "pool1"
+ pooling_param {
+ pool: MAX
+ kernel_size: 3
+ stride: 2
+ }
+}
+layer {
+ name: "relu1"
+ type: "ReLU"
+ bottom: "pool1"
+ top: "pool1"
+}
+layer {
+ name: "norm1"
+ type: "LRN"
+ bottom: "pool1"
+ top: "norm1"
+ lrn_param {
+ local_size: 3
+ alpha: 5e-05
+ beta: 0.75
+ norm_region: WITHIN_CHANNEL
+ }
+}
+layer {
+ name: "conv2"
+ type: "Convolution"
+ bottom: "norm1"
+ top: "conv2"
+ param {
+ lr_mult: 1
+ }
+ param {
+ lr_mult: 2
+ }
+ convolution_param {
+ num_output: 32
+ pad: 2
+ kernel_size: 5
+ stride: 1
+ weight_filler {
+ type: "gaussian"
+ std: 0.01
+ }
+ bias_filler {
+ type: "constant"
+ }
+ }
+}
+layer {
+ name: "relu2"
+ type: "ReLU"
+ bottom: "conv2"
+ top: "conv2"
+}
+layer {
+ name: "pool2"
+ type: "Pooling"
+ bottom: "conv2"
+ top: "pool2"
+ pooling_param {
+ pool: AVE
+ kernel_size: 3
+ stride: 2
+ }
+}
+layer {
+ name: "norm2"
+ type: "LRN"
+ bottom: "pool2"
+ top: "norm2"
+ lrn_param {
+ local_size: 3
+ alpha: 5e-05
+ beta: 0.75
+ norm_region: WITHIN_CHANNEL
+ }
+}
+layer {
+ name: "conv3"
+ type: "Convolution"
+ bottom: "norm2"
+ top: "conv3"
+ convolution_param {
+ num_output: 64
+ pad: 2
+ kernel_size: 5
+ stride: 1
+ weight_filler {
+ type: "gaussian"
+ std: 0.01
+ }
+ bias_filler {
+ type: "constant"
+ }
+ }
+}
+layer {
+ name: "relu3"
+ type: "ReLU"
+ bottom: "conv3"
+ top: "conv3"
+}
+layer {
+ name: "pool3"
+ type: "Pooling"
+ bottom: "conv3"
+ top: "pool3"
+ pooling_param {
+ pool: AVE
+ kernel_size: 3
+ stride: 2
+ }
+}
+layer {
+ name: "ip1"
+ type: "InnerProduct"
+ bottom: "pool3"
+ top: "ip1"
+ param {
+ lr_mult: 1
+ decay_mult: 250
+ }
+ param {
+ lr_mult: 2
+ decay_mult: 0
+ }
+ inner_product_param {
+ num_output: 10
+ weight_filler {
+ type: "gaussian"
+ std: 0.01
+ }
+ bias_filler {
+ type: "constant"
+ }
+ }
+}
+layer {
+ name: "accuracy"
+ type: "Accuracy"
+ bottom: "ip1"
+ bottom: "label"
+ top: "accuracy"
+ include {
+ phase: TEST
+ }
+}
+layer {
+ name: "loss"
+ type: "SoftmaxWithLoss"
+ bottom: "ip1"
+ bottom: "label"
+ top: "loss"
+}
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/examples/cifar10/caffe/cifar10_quick_solver.prototxt
----------------------------------------------------------------------
diff --git a/examples/cifar10/caffe/cifar10_quick_solver.prototxt b/examples/cifar10/caffe/cifar10_quick_solver.prototxt
new file mode 100644
index 0000000..3c5ce96
--- /dev/null
+++ b/examples/cifar10/caffe/cifar10_quick_solver.prototxt
@@ -0,0 +1,29 @@
+# From caffe repository
+# Commits on Sep 2, 2015
+
+# reduce the learning rate after 8 epochs (4000 iters) by a factor of 10
+
+# The train/test net protocol buffer definition
+net: "examples/cifar10/cifar10_quick_train_test.prototxt"
+# test_iter specifies how many forward passes the test should carry out.
+# In the case of MNIST, we have test batch size 100 and 100 test iterations,
+# covering the full 10,000 testing images.
+test_iter: 100
+# Carry out testing every 500 training iterations.
+test_interval: 500
+# The base learning rate, momentum and the weight decay of the network.
+base_lr: 0.001
+momentum: 0.9
+weight_decay: 0.004
+# The learning rate policy
+lr_policy: "fixed"
+# Display every 100 iterations
+display: 100
+# The maximum number of iterations
+max_iter: 4000
+# snapshot intermediate results
+snapshot: 4000
+snapshot_format: HDF5
+snapshot_prefix: "examples/cifar10/cifar10_quick"
+# solver mode: CPU or GPU
+solver_mode: GPU
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/examples/cifar10/caffe/cifar10_quick_train_test.prototxt
----------------------------------------------------------------------
diff --git a/examples/cifar10/caffe/cifar10_quick_train_test.prototxt b/examples/cifar10/caffe/cifar10_quick_train_test.prototxt
new file mode 100644
index 0000000..f83ca4b
--- /dev/null
+++ b/examples/cifar10/caffe/cifar10_quick_train_test.prototxt
@@ -0,0 +1,225 @@
+# From caffe repository
+# Commits on Feb 6, 2015
+
+name: "CIFAR10_quick"
+layer {
+ name: "cifar"
+ type: "Data"
+ top: "data"
+ top: "label"
+ include {
+ phase: TRAIN
+ }
+ transform_param {
+ mean_file: "examples/cifar10/mean.binaryproto"
+ }
+ data_param {
+ source: "examples/cifar10/cifar10_train_lmdb"
+ batch_size: 100
+ backend: LMDB
+ }
+}
+layer {
+ name: "cifar"
+ type: "Data"
+ top: "data"
+ top: "label"
+ include {
+ phase: TEST
+ }
+ transform_param {
+ mean_file: "examples/cifar10/mean.binaryproto"
+ }
+ data_param {
+ source: "examples/cifar10/cifar10_test_lmdb"
+ batch_size: 100
+ backend: LMDB
+ }
+}
+layer {
+ name: "conv1"
+ type: "Convolution"
+ bottom: "data"
+ top: "conv1"
+ param {
+ lr_mult: 1
+ }
+ param {
+ lr_mult: 2
+ }
+ convolution_param {
+ num_output: 32
+ pad: 2
+ kernel_size: 5
+ stride: 1
+ weight_filler {
+ type: "gaussian"
+ std: 0.0001
+ }
+ bias_filler {
+ type: "constant"
+ }
+ }
+}
+layer {
+ name: "pool1"
+ type: "Pooling"
+ bottom: "conv1"
+ top: "pool1"
+ pooling_param {
+ pool: MAX
+ kernel_size: 3
+ stride: 2
+ }
+}
+layer {
+ name: "relu1"
+ type: "ReLU"
+ bottom: "pool1"
+ top: "pool1"
+}
+layer {
+ name: "conv2"
+ type: "Convolution"
+ bottom: "pool1"
+ top: "conv2"
+ param {
+ lr_mult: 1
+ }
+ param {
+ lr_mult: 2
+ }
+ convolution_param {
+ num_output: 32
+ pad: 2
+ kernel_size: 5
+ stride: 1
+ weight_filler {
+ type: "gaussian"
+ std: 0.01
+ }
+ bias_filler {
+ type: "constant"
+ }
+ }
+}
+layer {
+ name: "relu2"
+ type: "ReLU"
+ bottom: "conv2"
+ top: "conv2"
+}
+layer {
+ name: "pool2"
+ type: "Pooling"
+ bottom: "conv2"
+ top: "pool2"
+ pooling_param {
+ pool: AVE
+ kernel_size: 3
+ stride: 2
+ }
+}
+layer {
+ name: "conv3"
+ type: "Convolution"
+ bottom: "pool2"
+ top: "conv3"
+ param {
+ lr_mult: 1
+ }
+ param {
+ lr_mult: 2
+ }
+ convolution_param {
+ num_output: 64
+ pad: 2
+ kernel_size: 5
+ stride: 1
+ weight_filler {
+ type: "gaussian"
+ std: 0.01
+ }
+ bias_filler {
+ type: "constant"
+ }
+ }
+}
+layer {
+ name: "relu3"
+ type: "ReLU"
+ bottom: "conv3"
+ top: "conv3"
+}
+layer {
+ name: "pool3"
+ type: "Pooling"
+ bottom: "conv3"
+ top: "pool3"
+ pooling_param {
+ pool: AVE
+ kernel_size: 3
+ stride: 2
+ }
+}
+layer {
+ name: "ip1"
+ type: "InnerProduct"
+ bottom: "pool3"
+ top: "ip1"
+ param {
+ lr_mult: 1
+ }
+ param {
+ lr_mult: 2
+ }
+ inner_product_param {
+ num_output: 64
+ weight_filler {
+ type: "gaussian"
+ std: 0.1
+ }
+ bias_filler {
+ type: "constant"
+ }
+ }
+}
+layer {
+ name: "ip2"
+ type: "InnerProduct"
+ bottom: "ip1"
+ top: "ip2"
+ param {
+ lr_mult: 1
+ }
+ param {
+ lr_mult: 2
+ }
+ inner_product_param {
+ num_output: 10
+ weight_filler {
+ type: "gaussian"
+ std: 0.1
+ }
+ bias_filler {
+ type: "constant"
+ }
+ }
+}
+layer {
+ name: "accuracy"
+ type: "Accuracy"
+ bottom: "ip2"
+ bottom: "label"
+ top: "accuracy"
+ include {
+ phase: TEST
+ }
+}
+layer {
+ name: "loss"
+ type: "SoftmaxWithLoss"
+ bottom: "ip2"
+ bottom: "label"
+ top: "loss"
+}
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/examples/cifar10/train.py
----------------------------------------------------------------------
diff --git a/examples/cifar10/train.py b/examples/cifar10/train.py
index 7494b8b..6443cf4 100644
--- a/examples/cifar10/train.py
+++ b/examples/cifar10/train.py
@@ -30,6 +30,7 @@ from singa import optimizer
from singa import device
from singa import tensor
from singa.proto import core_pb2
+from caffe import caffe_net
import alexnet
import vgg
@@ -105,6 +106,13 @@ def resnet_lr(epoch):
return 0.001
+def caffe_lr(epoch):
+ if epoch < 8:
+ return 0.001
+ else:
+ return 0.0001
+
+
def train(data, net, max_epoch, get_lr, weight_decay, batch_size=100,
use_cpu=False):
print 'Start intialization............'
@@ -163,8 +171,8 @@ def train(data, net, max_epoch, get_lr, weight_decay, batch_size=100,
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='Train dcnn for cifar10')
- parser.add_argument('model', choices=['vgg', 'alexnet', 'resnet'],
- default='alexnet')
+ parser.add_argument('model', choices=['vgg', 'alexnet', 'resnet', 'caffe'],
+ default='alexnet')
parser.add_argument('data', default='cifar-10-batches-py')
parser.add_argument('--use_cpu', action='store_true')
args = parser.parse_args()
@@ -173,10 +181,19 @@ if __name__ == '__main__':
print 'Loading data ..................'
train_x, train_y = load_train_data(args.data)
test_x, test_y = load_test_data(args.data)
- if args.model == 'alexnet':
+ if args.model == 'caffe':
+ train_x, test_x = normalize_for_alexnet(train_x, test_x)
+ net = caffe_net.create_net(args.use_cpu)
+ # for cifar10_full_train_test.prototxt
+ #train((train_x, train_y, test_x, test_y), net, 160, alexnet_lr, 0.004,
+ # use_cpu=args.use_cpu)
+ # for cifar10_quick_train_test.prototxt
+ train((train_x, train_y, test_x, test_y), net, 18, caffe_lr, 0.004,
+ use_cpu=args.use_cpu)
+ elif args.model == 'alexnet':
train_x, test_x = normalize_for_alexnet(train_x, test_x)
net = alexnet.create_net(args.use_cpu)
- train((train_x, train_y, test_x, test_y), net, 160, alexnet_lr, 0.004,
+ train((train_x, train_y, test_x, test_y), net, 2, alexnet_lr, 0.004,
use_cpu=args.use_cpu)
elif args.model == 'vgg':
train_x, test_x = normalize_for_vgg(train_x, test_x)
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/python/singa/converter.py
----------------------------------------------------------------------
diff --git a/python/singa/converter.py b/python/singa/converter.py
new file mode 100644
index 0000000..1378af0
--- /dev/null
+++ b/python/singa/converter.py
@@ -0,0 +1,155 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements. See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership. The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =============================================================================
+
+from google.protobuf import text_format
+from singa import layer
+from singa import metric
+from singa import loss
+from singa import net as ffnet
+from .proto import model_pb2
+from .proto import caffe_pb2
+
+
+class CaffeConverter:
+
+ def __init__(self, net_proto, solver_proto = None, input_sample_shape = None):
+ self.caffe_net_path = net_proto
+ self.caffe_solver_path = solver_proto
+ self.input_sample_shape = input_sample_shape
+
+ def read_net_proto(self):
+ net_config = caffe_pb2.NetParameter()
+ return self.read_proto(self.caffe_net_path, net_config)
+
+ def read_solver_proto(self):
+ solver_config = caffe_pb2.SolverParameter()
+ return self.read_proto(self.caffe_solver_path, solver_config)
+
+ def read_proto(self, filepath, parser_object):
+ file = open(filepath, "r")
+ if not file:
+ raise self.ProcessException("ERROR (" + filepath + ")!")
+ # Merges an ASCII representation of a protocol message into a message.
+ text_format.Merge(str(file.read()), parser_object)
+ file.close()
+ return parser_object
+
+ def convert_engine(self, layer_conf, solver_mode):
+ '''
+ Convert caffe engine into singa engine
+ return:
+ a singa engine string
+ '''
+ caffe_engine = ''
+ singa_engine = ''
+
+ # if no 'engine' field in caffe proto, set engine to -1
+ if layer_conf.type == 'Convolution' or layer_conf.type == 4:
+ caffe_engine = layer_conf.convolution_param.engine
+ elif layer_conf.type == 'Pooling' or layer_conf.type == 17:
+ caffe_engine = layer_conf.pooling_param.engine
+ elif layer_conf.type == 'ReLU' or layer_conf.type == 18:
+ caffe_engine = layer_conf.relu_param.engine
+ elif layer_conf.type == 'Sigmoid' or layer_conf.type == 19:
+ caffe_engine = layer_conf.sigmoid_param.engine
+ elif layer_conf.type == 'TanH' or layer_conf.type == 23:
+ caffe_engine = layer_conf.tanh_param.engine
+ elif layer_conf.type == 'LRN' or layer_conf.type == 15:
+ caffe_engine = layer_conf.lrn_param.engine
+ elif layer_conf.type == 'Softmax' or layer_conf.type == 20:
+ caffe_engine = layer_conf.softmax_param.engine
+ elif layer_conf.type == 'InnerProduct' or layer_conf.type == 14:
+ caffe_engine = -1
+ elif layer_conf.type == 'Dropout' or layer_conf.type == 6:
+ caffe_engine = -1
+ elif layer_conf.type == 'Flatten' or layer_conf.type == 8:
+ caffe_engine = -1
+ else:
+ raise Exception('Unknown layer type: ' + layer_conf.type)
+
+ # caffe_engine: -1-no field; 0-DEFAULT; 1-CAFFE; 2-CUDNN
+ # solver_mode: 0-CPU; 1-GPU
+ if solver_mode == 1:
+ singa_engine = 'cudnn'
+ else:
+ if caffe_engine == 2:
+ raise Exception('engine and solver mode mismatch!')
+ else:
+ singa_engine = 'singacpp'
+
+ if ((layer_conf.type == 'InnerProduct' or layer_conf.type == 14) or \
+ (layer_conf.type == 'Flatten' or layer_conf.type == 8)) and \
+ singa_engine == 'cudnn':
+ singa_engine = 'singacuda'
+
+ return singa_engine
+
+
+ def create_net(self):
+ '''
+ Create singa net based on caffe proto files.
+ net_proto: caffe prototxt that describes net
+ solver_proto: caffe prototxt that describe solver
+ input_sample_shape: shape of input data tensor
+ return:
+ a FeedForwardNet object
+ '''
+ caffe_net = self.read_net_proto()
+ if self.caffe_solver_path is not None:
+ caffe_solver = self.read_solver_proto()
+ layer_confs = ''
+ flatten_id = 0
+
+ # If the net proto has the input shape
+ if len(caffe_net.input_dim) > 0:
+ self.input_sample_shape = caffe_net.input_dim
+ if len(caffe_net.layer):
+ layer_confs = caffe_net.layer
+ elif len(caffe_net.layers):
+ layer_confs = caffe_net.layers
+ else:
+ raise Exception('Invalid proto file!')
+
+ net = ffnet.FeedForwardNet()
+ for i in range(len(layer_confs)):
+ if layer_confs[i].type == 'Data' or layer_confs[i].type == 5:
+ continue
+ elif layer_confs[i].type == 'SoftmaxWithLoss' or layer_confs[i].type == 21:
+ net.loss = loss.SoftmaxCrossEntropy()
+ elif layer_confs[i].type == 'EuclideanLoss' or layer_confs[i].type == 7:
+ net.loss = loss.SquareError()
+ elif layer_confs[i].type == 'Accuracy' or layer_confs[i].type == 1:
+ net.metric = metric.Accuracy()
+ else:
+ strConf = layer_confs[i].SerializeToString()
+ conf = model_pb2.LayerConf()
+ conf.ParseFromString(strConf)
+ if caffe_solver:
+ layer.engine = self.convert_engine(
+ layer_confs[i], caffe_solver.solver_mode)
+ else:
+ layer.engine = self.convert_engine(layer_confs[i], 0)
+ lyr = layer.Layer(conf.name, conf)
+ if len(net.layers) == 0:
+ lyr.setup(self.input_sample_shape)
+ print lyr.name, lyr.get_output_sample_shape()
+ if layer_confs[i].type == 'InnerProduct' or layer_confs[i].type == 14:
+ net.add(layer.Flatten('flat' + str(flatten_id)))
+ flatten_id += 1
+ net.add(lyr)
+
+ return net
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/77405ec1/python/singa/layer.py
----------------------------------------------------------------------
diff --git a/python/singa/layer.py b/python/singa/layer.py
index a22af55..e376bbf 100644
--- a/python/singa/layer.py
+++ b/python/singa/layer.py
@@ -78,14 +78,52 @@ class Layer(object):
name (str): layer name
'''
- def __init__(self, name, **kwargs):
- self.layer = None # layer converted by swig
- self.name = name # TODO(wangwei) duplicate with self.conf.name
- self.conf = model_pb2.LayerConf()
- self.conf.name = name
- self.param_specs = []
+ def __init__(self, name, conf=None, **kwargs):
+ if conf == None:
+ self.layer = None # layer converted by swig
+ self.name = name # TODO(wangwei) duplicate with self.conf.name
+ self.conf = model_pb2.LayerConf()
+ self.conf.name = name
+ self.param_specs = []
+ else:
+ self.conf = conf
+ self.name = str(conf.name)
+ self.caffe_layer()
+ self.param_specs = []
+
+ # convert caffe proto into singa proto format
+ # case1: parameters of conv and dense layers
+ # case2: type of activation layers
+ if (conf.type == 'Convolution' or conf.type == 4) or \
+ (conf.type == 'InnerProduct' or conf.type == 14):
+ w, b = _construct_param_specs_from_caffe_proto(conf)
+ del conf.param[:]
+ conf.param.extend([w, b])
+ self.param_specs.append(w)
+ self.param_specs.append(b)
+ #print 'conf:\n', conf
+ #if conf.type == 'Pooling':
+ # print 'conf:\n', conf
+
+ elif (conf.type == 'ReLU' or conf.type == 18) or \
+ (conf.type == 'Sigmoid' or conf.type == 19) or \
+ (conf.type == 'TanH' or conf.type == 23):
+ conf.type = (engine + '_' + conf.type).lower()
+ self.conf = conf
+
self.has_setup = False
+ def caffe_layer(self):
+ '''
+ Create a singa layer based on caffe layer configuration.
+ '''
+ _check_engine(engine, ['cudnn', 'singacpp', 'singacuda'])
+ if self.conf.type == 'InnerProduct' or self.conf.type == 14:
+ self.layer = _create_layer(engine, 'Dense')
+ else:
+ self.layer = _create_layer(engine, str(self.conf.type))
+
+
def param_names(self):
'''
Returns:
@@ -243,6 +281,7 @@ class Conv2D(Layer):
without the batchsize, e.g., (channel, height, width) or
(height, width, channel)
"""
+
def __init__(self, name, nb_kernels, kernel=3, stride=1, border_mode='same',
cudnn_prefer='fatest', data_format='NCHW',
use_bias=True, W_specs=None, b_specs=None,
@@ -319,6 +358,7 @@ class Pooling2D(Layer):
model_pb2.PoolingConf.AVE
'''
+
def __init__(self, name, mode, kernel=3, stride=2, border_mode='same',
pad=None, data_format='NCHW', input_sample_shape=None):
super(Pooling2D, self).__init__(name)
@@ -421,6 +461,7 @@ class BatchNormalization(Layer):
name (string): layer name
input_sample_shape (tuple): with at least one integer
"""
+
def __init__(self, name, momentum=0.9,
beta_specs=None, gamma_specs=None, input_sample_shape=None):
super(BatchNormalization, self).__init__(name)
@@ -434,8 +475,8 @@ class BatchNormalization(Layer):
beta_specs['name'] = name + '_beta'
if 'name' not in gamma_specs:
gamma_specs['name'] = name + '_gamma'
- mean_specs = {'init': 'constant', 'value': 0, 'name': name+'_mean'}
- var_specs = {'init': 'constant', 'value': 1, 'name': name+'_var'}
+ mean_specs = {'init': 'constant', 'value': 0, 'name': name + '_mean'}
+ var_specs = {'init': 'constant', 'value': 1, 'name': name + '_var'}
self.conf.param.extend([_construct_param_specs_from_dict(gamma_specs)])
self.conf.param.extend([_construct_param_specs_from_dict(beta_specs)])
self.conf.param.extend([_construct_param_specs_from_dict(mean_specs)])
@@ -499,6 +540,7 @@ class Dense(Layer):
W_transpose (bool): if true, output=x*W.T+b;
input_sample_shape (tuple): input feature length
"""
+
def __init__(self, name, num_output, use_bias=True,
W_specs=None, b_specs=None,
W_transpose=False, input_sample_shape=None):
@@ -579,6 +621,7 @@ class Activation(Layer):
mode (string): 'relu', 'sigmoid', or 'tanh'
input_sample_shape (tuple): shape of a single sample
"""
+
def __init__(self, name, mode='relu', input_sample_shape=None):
super(Activation, self).__init__(name)
_check_engine(engine, ['cudnn', 'singacpp', 'singacuda', 'singacl'])
@@ -596,6 +639,7 @@ class Softmax(Layer):
[0,axis) as the row, the [axis, -1) as the column.
input_sample_shape (tuple): shape of a single sample
"""
+
def __init__(self, name, axis=1, input_sample_shape=None):
super(Softmax, self).__init__(name)
# conf = self.conf.softmax_conf
@@ -615,6 +659,7 @@ class Flatten(Layer):
[0,axis) as the row, the [axis, -1) as the column.
input_sample_shape (tuple): shape for a single sample
"""
+
def __init__(self, name, axis=1, input_sample_shape=None):
super(Flatten, self).__init__(name)
conf = self.conf.flatten_conf
@@ -635,6 +680,7 @@ class Merge(Layer):
input_sample_shape: sample shape of the input. The sample shape of all
inputs should be the same.
'''
+
def __init__(self, name, input_sample_shape=None):
self.in_shape = input_sample_shape
self.num_input = 1
@@ -669,6 +715,7 @@ class Split(Layer):
input_sample_shape: includes a single integer for the input sample
feature size.
'''
+
def __init__(self, name, num_output, input_sample_shape=None):
self.num_output = num_output
self.in_shape = input_sample_shape
@@ -806,6 +853,7 @@ class RNN(Layer):
class LSTM(RNN):
+
def __init__(self, name, hidden_size, dropout=0.0, num_stacks=1,
input_mode='linear', bidirectional=False,
param_specs=None, input_sample_shape=None):
@@ -815,6 +863,7 @@ class LSTM(RNN):
class GRU(RNN):
+
def __init__(self, name, hidden_size, dropout=0.0, num_stacks=1,
input_mode='linear', bidirectional=False, param_specs=None,
input_sample_shape=None):
@@ -825,8 +874,8 @@ class GRU(RNN):
def _check_engine(engine, allowed_engines):
assert engine.lower() in Set(allowed_engines), \
- '%s is not a supported engine. Pls use one of %s' % \
- (engine, ', '.join(allowed_engines))
+ '%s is not a supported engine. Pls use one of %s' % \
+ (engine, ', '.join(allowed_engines))
def _create_layer(eng, layer):
@@ -839,6 +888,7 @@ def _create_layer(eng, layer):
layers, use the specific activation mode, e.g. 'relu', 'tanh'.
'''
layer_type = eng + '_' + layer
+ print layer_type.lower()
return singa_wrap.CreateLayer(layer_type.lower())
@@ -927,6 +977,63 @@ def _construct_param_specs_from_dict(specs):
return conf
+def _construct_param_specs_from_caffe_proto(lyr_conf):
+ """convert the param specs from a caffe layer proto into a singa paramspec
+ protobuf object.
+
+ args:
+ specs (dict): the fields inlcude
+ 'name' for parameter name
+ 'lr_mult' for learning rate multiplier;
+ 'decay_mult' for weight decay multiplier;
+ 'init' for init method, which could be 'gaussian', 'uniform',
+ 'xavier' and 'msra';
+ 'std', 'mean', 'high', 'low' are used by corresponding init methods;
+ caffe model has no 'constraint' and 'regularizer'
+
+ returns:
+ a pair of paramspec objects(weight and bias)
+ """
+ wparam = model_pb2.ParamSpec()
+ bparam = model_pb2.ParamSpec()
+ if len(lyr_conf.param) > 0:
+ wparam.name = lyr_conf.param[0].name
+ wparam.lr_mult = lyr_conf.param[0].lr_mult
+ wparam.decay_mult = lyr_conf.param[0].decay_mult
+ if len(lyr_conf.param) > 1:
+ bparam.name = lyr_conf.param[1].name
+ bparam.lr_mult = lyr_conf.param[1].lr_mult
+ bparam.decay_mult = lyr_conf.param[1].decay_mult
+ if wparam.name == '' or wparam.name is None:
+ wparam.name = lyr_conf.name + '_weight'
+ if bparam.name == '' or bparam.name is None:
+ bparam.name = lyr_conf.name + '_bias'
+ wfiller = wparam.filler
+ bfiller = bparam.filler
+ param = ''
+ if lyr_conf.type == 'Convolution' or lyr_conf.type == 4:
+ param = lyr_conf.convolution_conf
+ elif lyr_conf.type == 'InnerProduct' or lyr_conf.type == 14:
+ param = lyr_conf.dense_conf
+
+ if param != '':
+ wfiller.type = param.weight_filler.type.lower()
+ wfiller.min = param.weight_filler.min
+ wfiller.max = param.weight_filler.max
+ wfiller.mean = param.weight_filler.mean
+ wfiller.std = param.weight_filler.std
+ wfiller.value = param.weight_filler.value
+
+ bfiller.type = param.bias_filler.type.lower()
+ bfiller.min = param.bias_filler.min
+ bfiller.max = param.bias_filler.max
+ bfiller.mean = param.bias_filler.mean
+ bfiller.std = param.bias_filler.std
+ bfiller.value = param.bias_filler.value
+
+ return (wparam, bparam)
+
+
def get_layer_list():
""" Return a list of strings which include the identifiers (tags) of all
supported layers
[3/3] incubator-singa git commit: SINGA-253 Net converter for caffe
model
Posted by wa...@apache.org.
SINGA-253 Net converter for caffe model
Modification of pooling layer
add a 'ceil' field in proto to be compatible with caffe
train alexnet from caffe, achieving 82% accuracy.
Project: http://git-wip-us.apache.org/repos/asf/incubator-singa/repo
Commit: http://git-wip-us.apache.org/repos/asf/incubator-singa/commit/32b4368f
Tree: http://git-wip-us.apache.org/repos/asf/incubator-singa/tree/32b4368f
Diff: http://git-wip-us.apache.org/repos/asf/incubator-singa/diff/32b4368f
Branch: refs/heads/master
Commit: 32b4368f07b4c55880fa04bcc2d9f39ab9d75fc5
Parents: 77405ec
Author: Xiangrui <ca...@gmail.com>
Authored: Tue Oct 18 19:59:38 2016 +0800
Committer: Xiangrui <ca...@gmail.com>
Committed: Tue Oct 18 19:59:38 2016 +0800
----------------------------------------------------------------------
examples/cifar10/caffe/caffe_net.py | 8 ++++----
examples/cifar10/train.py | 8 ++++----
python/singa/layer.py | 6 +++---
src/model/layer/pooling.cc | 21 ++++++++++++++++-----
src/proto/model.proto | 5 ++++-
5 files changed, 31 insertions(+), 17 deletions(-)
----------------------------------------------------------------------
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/32b4368f/examples/cifar10/caffe/caffe_net.py
----------------------------------------------------------------------
diff --git a/examples/cifar10/caffe/caffe_net.py b/examples/cifar10/caffe/caffe_net.py
index 543e3a5..601fab2 100644
--- a/examples/cifar10/caffe/caffe_net.py
+++ b/examples/cifar10/caffe/caffe_net.py
@@ -23,10 +23,10 @@ def create_net(use_cpu):
if use_cpu:
layer.engine = 'singacpp'
- #net_proto = os.path.abspath('./caffe/cifar10_full_train_test.prototxt')
- #solver_proto = os.path.abspath('./caffe/cifar10_full_solver.prototxt')
- net_proto = os.path.abspath('./caffe/cifar10_quick_train_test.prototxt')
- solver_proto = os.path.abspath('./caffe/cifar10_quick_solver.prototxt')
+ net_proto = os.path.abspath('./caffe/cifar10_full_train_test.prototxt')
+ solver_proto = os.path.abspath('./caffe/cifar10_full_solver.prototxt')
+ #net_proto = os.path.abspath('./caffe/cifar10_quick_train_test.prototxt')
+ #solver_proto = os.path.abspath('./caffe/cifar10_quick_solver.prototxt')
input_sample_shape = [3, 32, 32, ]
cvt = converter.CaffeConverter(net_proto, solver_proto, input_sample_shape)
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/32b4368f/examples/cifar10/train.py
----------------------------------------------------------------------
diff --git a/examples/cifar10/train.py b/examples/cifar10/train.py
index 6443cf4..d54d694 100644
--- a/examples/cifar10/train.py
+++ b/examples/cifar10/train.py
@@ -185,11 +185,11 @@ if __name__ == '__main__':
train_x, test_x = normalize_for_alexnet(train_x, test_x)
net = caffe_net.create_net(args.use_cpu)
# for cifar10_full_train_test.prototxt
- #train((train_x, train_y, test_x, test_y), net, 160, alexnet_lr, 0.004,
- # use_cpu=args.use_cpu)
- # for cifar10_quick_train_test.prototxt
- train((train_x, train_y, test_x, test_y), net, 18, caffe_lr, 0.004,
+ train((train_x, train_y, test_x, test_y), net, 160, alexnet_lr, 0.004,
use_cpu=args.use_cpu)
+ # for cifar10_quick_train_test.prototxt
+ #train((train_x, train_y, test_x, test_y), net, 18, caffe_lr, 0.004,
+ # use_cpu=args.use_cpu)
elif args.model == 'alexnet':
train_x, test_x = normalize_for_alexnet(train_x, test_x)
net = alexnet.create_net(args.use_cpu)
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/32b4368f/python/singa/layer.py
----------------------------------------------------------------------
diff --git a/python/singa/layer.py b/python/singa/layer.py
index e376bbf..c7f0ce8 100644
--- a/python/singa/layer.py
+++ b/python/singa/layer.py
@@ -102,8 +102,9 @@ class Layer(object):
self.param_specs.append(w)
self.param_specs.append(b)
#print 'conf:\n', conf
- #if conf.type == 'Pooling':
- # print 'conf:\n', conf
+ if conf.type == 'Pooling':
+ conf.pooling_conf.ceil = True
+ #print 'conf:\n', conf
elif (conf.type == 'ReLU' or conf.type == 18) or \
(conf.type == 'Sigmoid' or conf.type == 19) or \
@@ -888,7 +889,6 @@ def _create_layer(eng, layer):
layers, use the specific activation mode, e.g. 'relu', 'tanh'.
'''
layer_type = eng + '_' + layer
- print layer_type.lower()
return singa_wrap.CreateLayer(layer_type.lower())
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/32b4368f/src/model/layer/pooling.cc
----------------------------------------------------------------------
diff --git a/src/model/layer/pooling.cc b/src/model/layer/pooling.cc
index 2ddfc1c..75dc133 100644
--- a/src/model/layer/pooling.cc
+++ b/src/model/layer/pooling.cc
@@ -18,6 +18,8 @@
#include "./pooling.h"
#include "singa/model/layer.h"
+#include <cmath>
+
namespace singa {
RegisterLayerClass(singacpp_pooling, Pooling);
@@ -62,11 +64,20 @@ void Pooling::Setup(const Shape& in_sample, const LayerConf& conf) {
height_ = in_sample.at(1);
width_ = in_sample.at(2);
pooled_height_ = 1;
- if (stride_h_ > 0)
- pooled_height_ =
- static_cast<size_t>((height_ + 2 * pad_h_ - kernel_h_) / stride_h_) + 1;
- pooled_width_ =
- static_cast<size_t>((width_ + 2 * pad_w_ - kernel_w_) / stride_w_) + 1;
+ if (pool_conf.ceil()) {
+ if (stride_h_ > 0)
+ pooled_height_ = static_cast<int>(ceil(static_cast<float>(
+ height_ + 2 * pad_h_ - kernel_h_) / stride_h_)) + 1;
+ pooled_width_ = static_cast<int>(ceil(static_cast<float>(
+ width_ + 2 * pad_w_ - kernel_w_) / stride_w_)) + 1;
+ }
+ else {
+ if (stride_h_ > 0)
+ pooled_height_ =
+ static_cast<size_t>((height_ + 2 * pad_h_ - kernel_h_) / stride_h_) + 1;
+ pooled_width_ =
+ static_cast<size_t>((width_ + 2 * pad_w_ - kernel_w_) / stride_w_) + 1;
+ }
out_sample_shape_ = vector<size_t>{channels_, pooled_height_, pooled_width_};
}
http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/32b4368f/src/proto/model.proto
----------------------------------------------------------------------
diff --git a/src/proto/model.proto b/src/proto/model.proto
index 641efb2..cfe946d 100644
--- a/src/proto/model.proto
+++ b/src/proto/model.proto
@@ -686,7 +686,10 @@ message PoolingConf {
optional bool global_pooling = 12 [default = false];
// whether to propagate nan
optional bool nan_prop = 53 [default = false];
-}
+
+ // Added by xiangrui, 18 Oct, 2016
+ optional bool ceil = 60 [default = false];
+}
message PowerConf {
// PowerLayer computes outputs y = (shift + scale * x) ^ power.