not sure

Joseph Redmon

2016-05-12 054e2b1954aafb15b0e983180dda309cfd5d831f

 src/parser.c

@@ -9,9 +9,11 @@
#include "convolutional_layer.h"
#include "activation_layer.h"
#include "normalization_layer.h"
#include "batchnorm_layer.h"
#include "deconvolutional_layer.h"
#include "connected_layer.h"
#include "rnn_layer.h"
#include "gru_layer.h"
#include "crnn_layer.h"
#include "maxpool_layer.h"
#include "softmax_layer.h"
@@ -37,12 +39,14 @@
int is_deconvolutional(section *s);
int is_connected(section *s);
int is_rnn(section *s);
int is_gru(section *s);
int is_crnn(section *s);
int is_maxpool(section *s);
int is_avgpool(section *s);
int is_dropout(section *s);
int is_softmax(section *s);
int is_normalization(section *s);
int is_batchnorm(section *s);
int is_crop(section *s);
int is_shortcut(section *s);
int is_cost(section *s);
@@ -157,8 +161,9 @@
    if(!(h && w && c)) error("Layer before convolutional layer must output image.");
    int batch_normalize = option_find_int_quiet(options, "batch_normalize", 0);
    int binary = option_find_int_quiet(options, "binary", 0);
    int xnor = option_find_int_quiet(options, "xnor", 0);

    convolutional_layer layer = make_convolutional_layer(batch,h,w,c,n,size,stride,pad,activation, batch_normalize, binary);
    convolutional_layer layer = make_convolutional_layer(batch,h,w,c,n,size,stride,pad,activation, batch_normalize, binary, xnor);
    layer.flipped = option_find_int_quiet(options, "flipped", 0);
    layer.dot = option_find_float_quiet(options, "dot", 0);

@@ -203,6 +208,16 @@
    return l;
}

layer parse_gru(list *options, size_params params)
{
    int output = option_find_int(options, "output",1);
    int batch_normalize = option_find_int_quiet(options, "batch_normalize", 0);

    layer l = make_gru_layer(params.batch, params.inputs, output, params.time_steps, batch_normalize);

    return l;
}

connected_layer parse_connected(list *options, size_params params)
{
    int output = option_find_int(options, "output",1);
@@ -333,6 +348,12 @@
    return l;
}

layer parse_batchnorm(list *options, size_params params)
{
    layer l = make_batchnorm_layer(params.batch, params.w, params.h, params.c);
    return l;
}

layer parse_shortcut(list *options, size_params params, network net)
{
    char *l = option_find(options, "from");   
@@ -411,6 +432,7 @@

learning_rate_policy get_policy(char *s)
{
    if (strcmp(s, "random")==0) return RANDOM;
    if (strcmp(s, "poly")==0) return POLY;
    if (strcmp(s, "constant")==0) return CONSTANT;
    if (strcmp(s, "step")==0) return STEP;
@@ -438,6 +460,7 @@
    net->c = option_find_int_quiet(options, "channels",0);
    net->inputs = option_find_int_quiet(options, "inputs", net->h * net->w * net->c);
    net->max_crop = option_find_int_quiet(options, "max_crop",net->w*2);
    net->min_crop = option_find_int_quiet(options, "min_crop",net->w);

    if(!net->inputs && !(net->h && net->w && net->c)) error("No input parameters supplied");

@@ -475,7 +498,7 @@
    } else if (net->policy == SIG){
        net->gamma = option_find_float(options, "gamma", 1);
        net->step = option_find_int(options, "step", 1);
    } else if (net->policy == POLY){
    } else if (net->policy == POLY || net->policy == RANDOM){
        net->power = option_find_float(options, "power", 1);
    }
    net->max_batches = option_find_int(options, "max_batches", 0);
@@ -520,6 +543,8 @@
            l = parse_deconvolutional(options, params);
        }else if(is_rnn(s)){
            l = parse_rnn(options, params);
        }else if(is_gru(s)){
            l = parse_gru(options, params);
        }else if(is_crnn(s)){
            l = parse_crnn(options, params);
        }else if(is_connected(s)){
@@ -534,6 +559,8 @@
            l = parse_softmax(options, params);
        }else if(is_normalization(s)){
            l = parse_normalization(options, params);
        }else if(is_batchnorm(s)){
            l = parse_batchnorm(options, params);
        }else if(is_maxpool(s)){
            l = parse_maxpool(options, params);
        }else if(is_avgpool(s)){
@@ -573,6 +600,40 @@
    return net;
}

LAYER_TYPE string_to_layer_type(char * type)
{

    if (strcmp(type, "[shortcut]")==0) return SHORTCUT;
    if (strcmp(type, "[crop]")==0) return CROP;
    if (strcmp(type, "[cost]")==0) return COST;
    if (strcmp(type, "[detection]")==0) return DETECTION;
    if (strcmp(type, "[local]")==0) return LOCAL;
    if (strcmp(type, "[deconv]")==0
            || strcmp(type, "[deconvolutional]")==0) return DECONVOLUTIONAL;
    if (strcmp(type, "[conv]")==0
            || strcmp(type, "[convolutional]")==0) return CONVOLUTIONAL;
    if (strcmp(type, "[activation]")==0) return ACTIVE;
    if (strcmp(type, "[net]")==0
            || strcmp(type, "[network]")==0) return NETWORK;
    if (strcmp(type, "[crnn]")==0) return CRNN;
    if (strcmp(type, "[gru]")==0) return GRU;
    if (strcmp(type, "[rnn]")==0) return RNN;
    if (strcmp(type, "[conn]")==0
            || strcmp(type, "[connected]")==0) return CONNECTED;
    if (strcmp(type, "[max]")==0
            || strcmp(type, "[maxpool]")==0) return MAXPOOL;
    if (strcmp(type, "[avg]")==0
            || strcmp(type, "[avgpool]")==0) return AVGPOOL;
    if (strcmp(type, "[dropout]")==0) return DROPOUT;
    if (strcmp(type, "[lrn]")==0
            || strcmp(type, "[normalization]")==0) return NORMALIZATION;
    if (strcmp(type, "[batchnorm]")==0) return BATCHNORM;
    if (strcmp(type, "[soft]")==0
            || strcmp(type, "[softmax]")==0) return SOFTMAX;
    if (strcmp(type, "[route]")==0) return ROUTE;
    return BLANK;
}

int is_shortcut(section *s)
{
    return (strcmp(s->type, "[shortcut]")==0);
@@ -616,6 +677,10 @@
{
    return (strcmp(s->type, "[crnn]")==0);
}
int is_gru(section *s)
{
    return (strcmp(s->type, "[gru]")==0);
}
int is_rnn(section *s)
{
    return (strcmp(s->type, "[rnn]")==0);
@@ -646,6 +711,11 @@
            || strcmp(s->type, "[normalization]")==0);
}

int is_batchnorm(section *s)
{
    return (strcmp(s->type, "[batchnorm]")==0);
}

int is_softmax(section *s)
{
    return (strcmp(s->type, "[soft]")==0
@@ -783,6 +853,18 @@
    fwrite(l.filters, sizeof(float), num, fp);
}

void save_batchnorm_weights(layer l, FILE *fp)
{
#ifdef GPU
    if(gpu_index >= 0){
        pull_batchnorm_layer(l);
    }
#endif
    fwrite(l.scales, sizeof(float), l.c, fp);
    fwrite(l.rolling_mean, sizeof(float), l.c, fp);
    fwrite(l.rolling_variance, sizeof(float), l.c, fp);
}

void save_connected_weights(layer l, FILE *fp)
{
#ifdef GPU
@@ -820,10 +902,19 @@
            save_convolutional_weights(l, fp);
        } if(l.type == CONNECTED){
            save_connected_weights(l, fp);
        } if(l.type == BATCHNORM){
            save_batchnorm_weights(l, fp);
        } if(l.type == RNN){
            save_connected_weights(*(l.input_layer), fp);
            save_connected_weights(*(l.self_layer), fp);
            save_connected_weights(*(l.output_layer), fp);
        } if(l.type == GRU){
            save_connected_weights(*(l.input_z_layer), fp);
            save_connected_weights(*(l.input_r_layer), fp);
            save_connected_weights(*(l.input_h_layer), fp);
            save_connected_weights(*(l.state_z_layer), fp);
            save_connected_weights(*(l.state_r_layer), fp);
            save_connected_weights(*(l.state_h_layer), fp);
        } if(l.type == CRNN){
            save_convolutional_weights(*(l.input_layer), fp);
            save_convolutional_weights(*(l.self_layer), fp);
@@ -867,10 +958,15 @@
    if(transpose){
        transpose_matrix(l.weights, l.inputs, l.outputs);
    }
    //printf("Biases: %f mean %f variance\n", mean_array(l.biases, l.outputs), variance_array(l.biases, l.outputs));
    //printf("Weights: %f mean %f variance\n", mean_array(l.weights, l.outputs*l.inputs), variance_array(l.weights, l.outputs*l.inputs));
    if (l.batch_normalize && (!l.dontloadscales)){
        fread(l.scales, sizeof(float), l.outputs, fp);
        fread(l.rolling_mean, sizeof(float), l.outputs, fp);
        fread(l.rolling_variance, sizeof(float), l.outputs, fp);
        //printf("Scales: %f mean %f variance\n", mean_array(l.scales, l.outputs), variance_array(l.scales, l.outputs));
        //printf("rolling_mean: %f mean %f variance\n", mean_array(l.rolling_mean, l.outputs), variance_array(l.rolling_mean, l.outputs));
        //printf("rolling_variance: %f mean %f variance\n", mean_array(l.rolling_variance, l.outputs), variance_array(l.rolling_variance, l.outputs));
    }
#ifdef GPU
    if(gpu_index >= 0){
@@ -879,6 +975,18 @@
#endif
}

void load_batchnorm_weights(layer l, FILE *fp)
{
    fread(l.scales, sizeof(float), l.c, fp);
    fread(l.rolling_mean, sizeof(float), l.c, fp);
    fread(l.rolling_variance, sizeof(float), l.c, fp);
#ifdef GPU
    if(gpu_index >= 0){
        push_batchnorm_layer(l);
    }
#endif
}

void load_convolutional_weights_binary(layer l, FILE *fp)
{
    fread(l.biases, sizeof(float), l.n, fp);
@@ -972,6 +1080,9 @@
        if(l.type == CONNECTED){
            load_connected_weights(l, fp, transpose);
        }
        if(l.type == BATCHNORM){
            load_batchnorm_weights(l, fp);
        }
        if(l.type == CRNN){
            load_convolutional_weights(*(l.input_layer), fp);
            load_convolutional_weights(*(l.self_layer), fp);
@@ -982,6 +1093,14 @@
            load_connected_weights(*(l.self_layer), fp, transpose);
            load_connected_weights(*(l.output_layer), fp, transpose);
        }
        if(l.type == GRU){
            load_connected_weights(*(l.input_z_layer), fp, transpose);
            load_connected_weights(*(l.input_r_layer), fp, transpose);
            load_connected_weights(*(l.input_h_layer), fp, transpose);
            load_connected_weights(*(l.state_z_layer), fp, transpose);
            load_connected_weights(*(l.state_r_layer), fp, transpose);
            load_connected_weights(*(l.state_h_layer), fp, transpose);
        }
        if(l.type == LOCAL){
            int locations = l.out_w*l.out_h;
            int size = l.size*l.size*l.c*l.n*locations;