From a723e1c62a27aeb39aaf7fcdeb3beb4e89fba32d Mon Sep 17 00:00:00 2001
From: Alexey <AlexeyAB@users.noreply.github.com>
Date: Wed, 15 Aug 2018 20:52:09 +0000
Subject: [PATCH] Merge pull request #766 from HotChick91/AlexeyAB-mask
---
src/convolutional_layer.c | 633 ++++++++++++++++++++++++++++++++++++++++++++++-----------
1 files changed, 507 insertions(+), 126 deletions(-)
diff --git a/src/convolutional_layer.c b/src/convolutional_layer.c
index ad2d8a5..e7c8e3f 100644
--- a/src/convolutional_layer.c
+++ b/src/convolutional_layer.c
@@ -8,6 +8,10 @@
#include <stdio.h>
#include <time.h>
+#ifdef CUDNN
+#pragma comment(lib, "cudnn.lib")
+#endif
+
#ifdef AI2
#include "xnor_layer.h"
#endif
@@ -19,28 +23,28 @@
void swap_binary(convolutional_layer *l)
{
- float *swap = l->filters;
- l->filters = l->binary_filters;
- l->binary_filters = swap;
+ float *swap = l->weights;
+ l->weights = l->binary_weights;
+ l->binary_weights = swap;
#ifdef GPU
- swap = l->filters_gpu;
- l->filters_gpu = l->binary_filters_gpu;
- l->binary_filters_gpu = swap;
+ swap = l->weights_gpu;
+ l->weights_gpu = l->binary_weights_gpu;
+ l->binary_weights_gpu = swap;
#endif
}
-void binarize_filters(float *filters, int n, int size, float *binary)
+void binarize_weights(float *weights, int n, int size, float *binary)
{
int i, f;
for(f = 0; f < n; ++f){
float mean = 0;
for(i = 0; i < size; ++i){
- mean += fabs(filters[f*size + i]);
+ mean += fabs(weights[f*size + i]);
}
mean = mean / size;
for(i = 0; i < size; ++i){
- binary[f*size + i] = (filters[f*size + i] > 0) ? mean : -mean;
+ binary[f*size + i] = (weights[f*size + i] > 0) ? mean: -mean;
}
}
}
@@ -103,7 +107,7 @@
size_t s = 0;
cudnnGetConvolutionForwardWorkspaceSize(cudnn_handle(),
l.srcTensorDesc,
- l.filterDesc,
+ l.weightDesc,
l.convDesc,
l.dstTensorDesc,
l.fw_algo,
@@ -113,12 +117,12 @@
l.srcTensorDesc,
l.ddstTensorDesc,
l.convDesc,
- l.dfilterDesc,
+ l.dweightDesc,
l.bf_algo,
&s);
if (s > most) most = s;
cudnnGetConvolutionBackwardDataWorkspaceSize(cudnn_handle(),
- l.filterDesc,
+ l.weightDesc,
l.ddstTensorDesc,
l.convDesc,
l.dsrcTensorDesc,
@@ -133,45 +137,126 @@
#ifdef GPU
#ifdef CUDNN
-void cudnn_convolutional_setup(layer *l)
+void cudnn_convolutional_setup(layer *l, int cudnn_preference)
{
- cudnnSetTensor4dDescriptor(l->dsrcTensorDesc, CUDNN_TENSOR_NCHW, CUDNN_DATA_FLOAT, l->batch, l->c, l->h, l->w);
- cudnnSetTensor4dDescriptor(l->ddstTensorDesc, CUDNN_TENSOR_NCHW, CUDNN_DATA_FLOAT, l->batch, l->out_c, l->out_h, l->out_w);
- cudnnSetFilter4dDescriptor(l->dfilterDesc, CUDNN_DATA_FLOAT, CUDNN_TENSOR_NCHW, l->n, l->c, l->size, l->size);
- cudnnSetTensor4dDescriptor(l->srcTensorDesc, CUDNN_TENSOR_NCHW, CUDNN_DATA_FLOAT, l->batch, l->c, l->h, l->w);
- cudnnSetTensor4dDescriptor(l->dstTensorDesc, CUDNN_TENSOR_NCHW, CUDNN_DATA_FLOAT, l->batch, l->out_c, l->out_h, l->out_w);
- cudnnSetFilter4dDescriptor(l->filterDesc, CUDNN_DATA_FLOAT, CUDNN_TENSOR_NCHW, l->n, l->c, l->size, l->size);
- cudnnSetConvolution2dDescriptor(l->convDesc, l->pad, l->pad, l->stride, l->stride, 1, 1, CUDNN_CROSS_CORRELATION);
+#ifdef CUDNN_HALF
+ // TRUE_HALF_CONFIG is only supported on architectures with true fp16 support (compute capability 5.3 and 6.0):
+ // Tegra X1, Jetson TX1, DRIVE CX, DRIVE PX, Quadro GP100, Tesla P100
+ // PSEUDO_HALF_CONFIG is required for Tensor Cores - our case!
+ const cudnnDataType_t data_type = CUDNN_DATA_HALF;
+#else
+ cudnnDataType_t data_type = CUDNN_DATA_FLOAT;
+#endif
+
+#if(CUDNN_MAJOR >= 7)
+ // Tensor Core uses CUDNN_TENSOR_OP_MATH instead of CUDNN_DEFAULT_MATH
+ // For *_ALGO_WINOGRAD_NONFUSED can be used CUDNN_DATA_FLOAT
+ // otherwise Input, Filter and Output descriptors (xDesc, yDesc, wDesc, dxDesc, dyDesc and dwDesc as applicable) have dataType = CUDNN_DATA_HALF
+ // Three techniques for training using Mixed-precision: https://devblogs.nvidia.com/mixed-precision-training-deep-neural-networks/
+ // 1. Accumulation into FP32
+ // 2. Loss Scaling - required only for: activation gradients. We do not use.
+ // 3. FP32 Master Copy of Weights
+ // More: http://docs.nvidia.com/deeplearning/sdk/cudnn-developer-guide/index.html#tensor_ops
+ cudnnSetConvolutionMathType(l->convDesc, CUDNN_TENSOR_OP_MATH);
+#endif
+
+ // INT8_CONFIG, INT8_EXT_CONFIG, INT8x4_CONFIG and INT8x4_EXT_CONFIG are only supported
+ // on architectures with DP4A support (compute capability 6.1 and later).
+ //cudnnDataType_t data_type = CUDNN_DATA_INT8;
+
+ // backward delta
+ cudnnSetTensor4dDescriptor(l->dsrcTensorDesc, CUDNN_TENSOR_NCHW, data_type, l->batch, l->c, l->h, l->w);
+ cudnnSetTensor4dDescriptor(l->ddstTensorDesc, CUDNN_TENSOR_NCHW, data_type, l->batch, l->out_c, l->out_h, l->out_w);
+ cudnnSetFilter4dDescriptor(l->dweightDesc, data_type, CUDNN_TENSOR_NCHW, l->n, l->c, l->size, l->size);
+
+ // forward
+ cudnnSetTensor4dDescriptor(l->srcTensorDesc, CUDNN_TENSOR_NCHW, data_type, l->batch, l->c, l->h, l->w);
+ cudnnSetTensor4dDescriptor(l->dstTensorDesc, CUDNN_TENSOR_NCHW, data_type, l->batch, l->out_c, l->out_h, l->out_w);
+ cudnnSetFilter4dDescriptor(l->weightDesc, data_type, CUDNN_TENSOR_NCHW, l->n, l->c, l->size, l->size);
+
+ // batch norm
+ cudnnSetTensor4dDescriptor(l->normTensorDesc, CUDNN_TENSOR_NCHW, CUDNN_DATA_FLOAT, 1, l->out_c, 1, 1);
+ cudnnSetTensor4dDescriptor(l->normDstTensorDesc, CUDNN_TENSOR_NCHW, CUDNN_DATA_FLOAT, l->batch, l->out_c, l->out_h, l->out_w);
+
+ cudnnSetTensor4dDescriptor(l->normDstTensorDescF16, CUDNN_TENSOR_NCHW, data_type, l->batch, l->out_c, l->out_h, l->out_w);
+#if(CUDNN_MAJOR >= 6)
+ cudnnSetConvolution2dDescriptor(l->convDesc, l->pad, l->pad, l->stride, l->stride, 1, 1, CUDNN_CROSS_CORRELATION, CUDNN_DATA_FLOAT); // cudnn >= 6.0
+#else
+ cudnnSetConvolution2dDescriptor(l->convDesc, l->pad, l->pad, l->stride, l->stride, 1, 1, CUDNN_CROSS_CORRELATION); // cudnn 5.1
+#endif
+ int forward_algo = CUDNN_CONVOLUTION_FWD_PREFER_FASTEST;
+ int backward_algo = CUDNN_CONVOLUTION_BWD_DATA_PREFER_FASTEST;
+ int backward_filter = CUDNN_CONVOLUTION_BWD_FILTER_PREFER_FASTEST;
+ if (cudnn_preference == cudnn_smallest)
+ {
+ forward_algo = CUDNN_CONVOLUTION_FWD_NO_WORKSPACE;
+ backward_algo = CUDNN_CONVOLUTION_BWD_DATA_NO_WORKSPACE;
+ backward_filter = CUDNN_CONVOLUTION_BWD_FILTER_NO_WORKSPACE;
+ printf(" CUDNN-slow ");
+ }
+
cudnnGetConvolutionForwardAlgorithm(cudnn_handle(),
l->srcTensorDesc,
- l->filterDesc,
+ l->weightDesc,
l->convDesc,
l->dstTensorDesc,
- CUDNN_CONVOLUTION_FWD_PREFER_FASTEST,
+ forward_algo,
0,
&l->fw_algo);
cudnnGetConvolutionBackwardDataAlgorithm(cudnn_handle(),
- l->filterDesc,
+ l->weightDesc,
l->ddstTensorDesc,
l->convDesc,
l->dsrcTensorDesc,
- CUDNN_CONVOLUTION_BWD_DATA_PREFER_FASTEST,
+ backward_algo,
0,
&l->bd_algo);
cudnnGetConvolutionBackwardFilterAlgorithm(cudnn_handle(),
l->srcTensorDesc,
l->ddstTensorDesc,
l->convDesc,
- l->dfilterDesc,
- CUDNN_CONVOLUTION_BWD_FILTER_PREFER_FASTEST,
+ l->dweightDesc,
+ backward_filter,
0,
&l->bf_algo);
+
+ if (data_type == CUDNN_DATA_HALF)
+ {
+ // HALF-16 if(data_type == CUDNN_DATA_HALF)
+ l->fw_algo = CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM;
+ l->bd_algo = CUDNN_CONVOLUTION_BWD_DATA_ALGO_1;
+ l->bf_algo = CUDNN_CONVOLUTION_BWD_FILTER_ALGO_1;
+
+ // FLOAT-32 if(data_type == CUDNN_DATA_FLOAT)
+ //l->fw_algo = CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD_NONFUSED;
+ //l->bd_algo = CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD_NONFUSED;
+ //l->bf_algo = CUDNN_CONVOLUTION_BWD_FILTER_ALGO_WINOGRAD_NONFUSED;
+
+ int fw = 0, bd = 0, bf = 0;
+ if (l->fw_algo == CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM) fw = 1;
+ //printf("Tensor Cores - Forward enabled: l->fw_algo = CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM \n");
+ if (l->fw_algo == CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD_NONFUSED) fw = 2;
+ //printf("Tensor Cores - Forward enabled: l->fw_algo = CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD_NONFUSED \n");
+
+ if (l->bd_algo == CUDNN_CONVOLUTION_BWD_DATA_ALGO_1) bd = 1;
+ //printf("Tensor Cores - Backward-data enabled: l->bd_algo = CUDNN_CONVOLUTION_BWD_DATA_ALGO_1 \n");
+ if (l->bd_algo == CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD_NONFUSED) bd = 2;
+ //printf("Tensor Cores - Backward-data enabled: l->bd_algo = CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD_NONFUSED \n");
+
+ if (l->bf_algo == CUDNN_CONVOLUTION_BWD_FILTER_ALGO_1) bf = 1;
+ //printf("Tensor Cores - Backward-filter enabled: l->bf_algo = CUDNN_CONVOLUTION_BWD_FILTER_ALGO_1 \n");
+ if (l->bf_algo == CUDNN_CONVOLUTION_BWD_FILTER_ALGO_WINOGRAD_NONFUSED) bf = 2;
+ //printf("Tensor Cores - Backward-filter enabled: l->bf_algo = CUDNN_CONVOLUTION_BWD_FILTER_ALGO_WINOGRAD_NONFUSED \n");
+
+ //if (fw == 2 && bd == 2 && bf == 2) printf("TF ");
+ //else if (fw == 1 && bd == 1 && bf == 1) printf("TH ");
+ }
}
#endif
#endif
-convolutional_layer make_convolutional_layer(int batch, int h, int w, int c, int n, int size, int stride, int padding, ACTIVATION activation, int batch_normalize, int binary, int xnor)
+convolutional_layer make_convolutional_layer(int batch, int h, int w, int c, int n, int size, int stride, int padding, ACTIVATION activation, int batch_normalize, int binary, int xnor, int adam)
{
int i;
convolutional_layer l = {0};
@@ -189,15 +274,15 @@
l.pad = padding;
l.batch_normalize = batch_normalize;
- l.filters = calloc(c*n*size*size, sizeof(float));
- l.filter_updates = calloc(c*n*size*size, sizeof(float));
+ l.weights = calloc(c*n*size*size, sizeof(float));
+ l.weight_updates = calloc(c*n*size*size, sizeof(float));
l.biases = calloc(n, sizeof(float));
l.bias_updates = calloc(n, sizeof(float));
// float scale = 1./sqrt(size*size*c);
float scale = sqrt(2./(size*size*c));
- for(i = 0; i < c*n*size*size; ++i) l.filters[i] = scale*rand_uniform(-1, 1);
+ for(i = 0; i < c*n*size*size; ++i) l.weights[i] = scale*rand_uniform(-1, 1);
int out_h = convolutional_out_height(l);
int out_w = convolutional_out_width(l);
l.out_h = out_h;
@@ -206,16 +291,19 @@
l.outputs = l.out_h * l.out_w * l.out_c;
l.inputs = l.w * l.h * l.c;
- l.output = calloc(l.batch*out_h * out_w * n, sizeof(float));
- l.delta = calloc(l.batch*out_h * out_w * n, sizeof(float));
+ l.output = calloc(l.batch*l.outputs, sizeof(float));
+ l.delta = calloc(l.batch*l.outputs, sizeof(float));
+ l.forward = forward_convolutional_layer;
+ l.backward = backward_convolutional_layer;
+ l.update = update_convolutional_layer;
if(binary){
- l.binary_filters = calloc(c*n*size*size, sizeof(float));
- l.cfilters = calloc(c*n*size*size, sizeof(char));
+ l.binary_weights = calloc(c*n*size*size, sizeof(float));
+ l.cweights = calloc(c*n*size*size, sizeof(char));
l.scales = calloc(n, sizeof(float));
}
if(xnor){
- l.binary_filters = calloc(c*n*size*size, sizeof(float));
+ l.binary_weights = calloc(c*n*size*size, sizeof(float));
l.binary_input = calloc(l.inputs*l.batch, sizeof(float));
}
@@ -229,29 +317,49 @@
l.mean = calloc(n, sizeof(float));
l.variance = calloc(n, sizeof(float));
+ l.mean_delta = calloc(n, sizeof(float));
+ l.variance_delta = calloc(n, sizeof(float));
+
l.rolling_mean = calloc(n, sizeof(float));
l.rolling_variance = calloc(n, sizeof(float));
+ l.x = calloc(l.batch*l.outputs, sizeof(float));
+ l.x_norm = calloc(l.batch*l.outputs, sizeof(float));
+ }
+ if(adam){
+ l.adam = 1;
+ l.m = calloc(c*n*size*size, sizeof(float));
+ l.v = calloc(c*n*size*size, sizeof(float));
}
#ifdef GPU
+ l.forward_gpu = forward_convolutional_layer_gpu;
+ l.backward_gpu = backward_convolutional_layer_gpu;
+ l.update_gpu = update_convolutional_layer_gpu;
+
if(gpu_index >= 0){
- l.filters_gpu = cuda_make_array(l.filters, c*n*size*size);
- l.filter_updates_gpu = cuda_make_array(l.filter_updates, c*n*size*size);
+ if (adam) {
+ l.m_gpu = cuda_make_array(l.m, c*n*size*size);
+ l.v_gpu = cuda_make_array(l.v, c*n*size*size);
+ }
+
+ l.weights_gpu = cuda_make_array(l.weights, c*n*size*size);
+#ifdef CUDNN_HALF
+ l.weights_gpu16 = cuda_make_array(NULL, c*n*size*size / 2); //cuda_make_array(l.weights, c*n*size*size / 2);
+ l.weight_updates_gpu16 = cuda_make_array(NULL, c*n*size*size / 2); //cuda_make_array(l.weight_updates, c*n*size*size / 2);
+#endif
+ l.weight_updates_gpu = cuda_make_array(l.weight_updates, c*n*size*size);
l.biases_gpu = cuda_make_array(l.biases, n);
l.bias_updates_gpu = cuda_make_array(l.bias_updates, n);
- l.scales_gpu = cuda_make_array(l.scales, n);
- l.scale_updates_gpu = cuda_make_array(l.scale_updates, n);
-
l.delta_gpu = cuda_make_array(l.delta, l.batch*out_h*out_w*n);
l.output_gpu = cuda_make_array(l.output, l.batch*out_h*out_w*n);
if(binary){
- l.binary_filters_gpu = cuda_make_array(l.filters, c*n*size*size);
+ l.binary_weights_gpu = cuda_make_array(l.weights, c*n*size*size);
}
if(xnor){
- l.binary_filters_gpu = cuda_make_array(l.filters, c*n*size*size);
+ l.binary_weights_gpu = cuda_make_array(l.weights, c*n*size*size);
l.binary_input_gpu = cuda_make_array(0, l.inputs*l.batch);
}
@@ -265,25 +373,33 @@
l.mean_delta_gpu = cuda_make_array(l.mean, n);
l.variance_delta_gpu = cuda_make_array(l.variance, n);
+ l.scales_gpu = cuda_make_array(l.scales, n);
+ l.scale_updates_gpu = cuda_make_array(l.scale_updates, n);
+
l.x_gpu = cuda_make_array(l.output, l.batch*out_h*out_w*n);
l.x_norm_gpu = cuda_make_array(l.output, l.batch*out_h*out_w*n);
}
#ifdef CUDNN
+ cudnnCreateTensorDescriptor(&l.normDstTensorDesc);
+ cudnnCreateTensorDescriptor(&l.normDstTensorDescF16);
+ cudnnCreateTensorDescriptor(&l.normTensorDesc);
cudnnCreateTensorDescriptor(&l.srcTensorDesc);
cudnnCreateTensorDescriptor(&l.dstTensorDesc);
- cudnnCreateFilterDescriptor(&l.filterDesc);
+ cudnnCreateFilterDescriptor(&l.weightDesc);
cudnnCreateTensorDescriptor(&l.dsrcTensorDesc);
cudnnCreateTensorDescriptor(&l.ddstTensorDesc);
- cudnnCreateFilterDescriptor(&l.dfilterDesc);
+ cudnnCreateFilterDescriptor(&l.dweightDesc);
cudnnCreateConvolutionDescriptor(&l.convDesc);
- cudnn_convolutional_setup(&l);
+ cudnn_convolutional_setup(&l, cudnn_fastest);
#endif
}
#endif
l.workspace_size = get_workspace_size(l);
l.activation = activation;
- fprintf(stderr, "Convolutional Layer: %d x %d x %d image, %d filters -> %d x %d x %d image\n", h,w,c,n, out_h, out_w, n);
+ //fprintf(stderr, "conv %5d %2d x%2d /%2d %4d x%4d x%4d -> %4d x%4d x%4d\n", n, size, size, stride, w, h, c, l.out_w, l.out_h, l.out_c);
+ l.bflops = (2.0 * l.n * l.size*l.size*l.c * l.out_h*l.out_w) / 1000000000.;
+ fprintf(stderr, "conv %5d %2d x%2d /%2d %4d x%4d x%4d -> %4d x%4d x%4d %5.3f BF\n", n, size, size, stride, w, h, c, l.out_w, l.out_h, l.out_c, l.bflops);
return l;
}
@@ -294,7 +410,7 @@
for(i = 0; i < l.n; ++i){
float scale = l.scales[i]/sqrt(l.rolling_variance[i] + .00001);
for(j = 0; j < l.c*l.size*l.size; ++j){
- l.filters[i*l.c*l.size*l.size + j] *= scale;
+ l.weights[i*l.c*l.size*l.size + j] *= scale;
}
l.biases[i] -= l.rolling_mean[i] * scale;
l.scales[i] = 1;
@@ -305,7 +421,7 @@
void test_convolutional_layer()
{
- convolutional_layer l = make_convolutional_layer(1, 5, 5, 3, 2, 5, 2, 1, LEAKY, 1, 0, 0);
+ convolutional_layer l = make_convolutional_layer(1, 5, 5, 3, 2, 5, 2, 1, LEAKY, 1, 0, 0, 0);
l.batch_normalize = 1;
float data[] = {1,1,1,1,1,
1,1,1,1,1,
@@ -329,6 +445,8 @@
void resize_convolutional_layer(convolutional_layer *l, int w, int h)
{
+ int old_w = l->w;
+ int old_h = l->h;
l->w = w;
l->h = h;
int out_w = convolutional_out_width(*l);
@@ -340,22 +458,55 @@
l->outputs = l->out_h * l->out_w * l->out_c;
l->inputs = l->w * l->h * l->c;
- l->output = realloc(l->output,
- l->batch*out_h * out_w * l->n*sizeof(float));
- l->delta = realloc(l->delta,
- l->batch*out_h * out_w * l->n*sizeof(float));
+ l->output = realloc(l->output, l->batch*l->outputs*sizeof(float));
+ l->delta = realloc(l->delta, l->batch*l->outputs*sizeof(float));
+ if(l->batch_normalize){
+ l->x = realloc(l->x, l->batch*l->outputs*sizeof(float));
+ l->x_norm = realloc(l->x_norm, l->batch*l->outputs*sizeof(float));
+ }
+
+ if (l->xnor) {
+ //l->binary_input = realloc(l->inputs*l->batch, sizeof(float));
+ }
#ifdef GPU
- cuda_free(l->delta_gpu);
- cuda_free(l->output_gpu);
+ if (old_w < w || old_h < h) {
+ cuda_free(l->delta_gpu);
+ cuda_free(l->output_gpu);
- l->delta_gpu = cuda_make_array(l->delta, l->batch*out_h*out_w*l->n);
- l->output_gpu = cuda_make_array(l->output, l->batch*out_h*out_w*l->n);
+ l->delta_gpu = cuda_make_array(l->delta, l->batch*l->outputs);
+ l->output_gpu = cuda_make_array(l->output, l->batch*l->outputs);
+
+ if (l->batch_normalize) {
+ cuda_free(l->x_gpu);
+ cuda_free(l->x_norm_gpu);
+
+ l->x_gpu = cuda_make_array(l->output, l->batch*l->outputs);
+ l->x_norm_gpu = cuda_make_array(l->output, l->batch*l->outputs);
+ }
+
+ if (l->xnor) {
+ cuda_free(l->binary_input_gpu);
+ l->binary_input_gpu = cuda_make_array(0, l->inputs*l->batch);
+ }
+ }
#ifdef CUDNN
- cudnn_convolutional_setup(l);
+ cudnn_convolutional_setup(l, cudnn_fastest);
#endif
#endif
l->workspace_size = get_workspace_size(*l);
+
+#ifdef CUDNN
+ // check for excessive memory consumption
+ size_t free_byte;
+ size_t total_byte;
+ check_error(cudaMemGetInfo(&free_byte, &total_byte));
+ if (l->workspace_size > free_byte || l->workspace_size >= total_byte / 2) {
+ printf(" used slow CUDNN algo without Workspace! Need memory: %zu, available: %zu\n", l->workspace_size, (free_byte < total_byte/2) ? free_byte : total_byte/2);
+ cudnn_convolutional_setup(l, cudnn_smallest);
+ l->workspace_size = get_workspace_size(*l);
+ }
+#endif
}
void add_bias(float *output, float *biases, int batch, int n, int size)
@@ -392,49 +543,125 @@
}
}
+void gemm_nn_custom(int M, int N, int K, float ALPHA,
+ float *A, int lda,
+ float *B, int ldb,
+ float *C, int ldc)
+{
+ int i, j, k;
+ for (i = 0; i < M; ++i) {
+ for (k = 0; k < K; ++k) {
+ register float A_PART = ALPHA*A[i*lda + k];
+ //printf("\n weight = %f \n", A_PART);
+ for (j = 0; j < N; ++j) {
+ C[i*ldc + j] += A_PART*B[k*ldb + j];
+ }
+ }
+ }
+}
+
+
+void get_mean_array(float *src, size_t size, size_t filters, float *mean_arr) {
+ size_t i, counter;
+ counter = 0;
+ for (i = 0; i < size; i += size / filters) {
+ mean_arr[counter++] = fabs(src[i]);
+ }
+}
+
+/*
+void float_to_bit(float *src, unsigned char *dst, size_t size) {
+
+ size_t dst_size = size / 8 + 1;
+ memset(dst, 0, dst_size);
+ size_t i, dst_i, dst_shift;
+ for (i = 0; i < size; ++i) {
+ if (src[i] > 0) set_bit(dst, i);
+ }
+}
+*/
+
+void bit_to_float(unsigned char *src, float *dst, size_t size, size_t filters, float *mean_arr) {
+ memset(dst, 0, size *sizeof(float));
+ size_t i, src_i, src_shift;
+
+ for (i = 0; i < size; ++i) {
+ float mean_val = 1;
+ if(mean_arr != NULL) mean_val = fabs(mean_arr[i / (size / filters)]);
+ if(get_bit(src, i)) dst[i] = mean_val;
+ else dst[i] = -mean_val;
+ }
+}
+
+void binary_align_weights(convolutional_layer *l)
+{
+ int m = l->n;
+ int k = l->size*l->size*l->c;
+ size_t new_lda = k + (l->lda_align - k % l->lda_align); // (k / 8 + 1) * 8;
+
+ binarize_weights(l->weights, m, k, l->binary_weights);
+
+ size_t align_weights_size = new_lda * m;
+ size_t align_bit_weights_size = align_weights_size / 8;// +1;
+ float *align_weights = calloc(align_weights_size, sizeof(float));
+ l->align_bit_weights = calloc(align_bit_weights_size, sizeof(char));
+
+ size_t i, j;
+ // align A without transpose
+ for (i = 0; i < m; ++i) {
+ for (j = 0; j < k; ++j) {
+ align_weights[i*new_lda + j] = l->binary_weights[i*k + j];
+ }
+ }
+ float_to_bit(align_weights, l->align_bit_weights, align_weights_size);
+
+ l->mean_arr = calloc(l->n, sizeof(float));
+ get_mean_array(align_weights, align_weights_size, l->n, l->mean_arr);
+
+ free(align_weights);
+}
+
+// further optimizations: im2col_bin() for XNOR, and then transpose_aling_bin()
+size_t binary_transpose_align_input(int k, int n, float *b, char **t_bit_input, size_t ldb_align)
+{
+ size_t new_ldb = k + (ldb_align - k%ldb_align); // (k / 8 + 1) * 8;
+ size_t t_intput_size = new_ldb * n;
+ size_t t_bit_input_size = t_intput_size / 8;// +1;
+ float *t_input = calloc(t_intput_size, sizeof(float));
+ //char *
+ *t_bit_input = calloc(t_bit_input_size, sizeof(char));
+
+ //printf("\n bit_input_size = %d, n = %d, k = %d, ldb = %d \n", bit_input_size, n, k, n);
+ //printf("\n t_bit_input_size = %d, k = %d, n = %d, new_ldb = %d \n", t_bit_input_size, k, n, new_ldb);
+
+ //printf("\n align_weights_size = %d, k = %d, m = %d, lda = %d \n", align_weights_size, k, m, k);
+ //printf("\n align_bit_weights_size = %d, k = %d, m = %d, new_lda = %d \n", align_bit_weights_size, k, m, new_ldb);
+
+ int blocksize = 64;
+ transpose_block_SSE4x4(b, t_input, k, n, n, new_ldb, blocksize);
+
+ //printf("\n blocksize = %d \n", blocksize);
+
+ float_to_bit(t_input, *t_bit_input, t_intput_size);
+ free(t_input);
+
+ return t_intput_size;
+}
+
+
void forward_convolutional_layer(convolutional_layer l, network_state state)
{
int out_h = convolutional_out_height(l);
int out_w = convolutional_out_width(l);
int i;
-
fill_cpu(l.outputs*l.batch, 0, l.output, 1);
- /*
- if(l.binary){
- binarize_filters(l.filters, l.n, l.c*l.size*l.size, l.binary_filters);
- binarize_filters2(l.filters, l.n, l.c*l.size*l.size, l.cfilters, l.scales);
- swap_binary(&l);
- }
- */
-
- /*
- if(l.binary){
- int m = l.n;
- int k = l.size*l.size*l.c;
- int n = out_h*out_w;
-
- char *a = l.cfilters;
- float *b = state.workspace;
- float *c = l.output;
-
- for(i = 0; i < l.batch; ++i){
- im2col_cpu(state.input, l.c, l.h, l.w,
- l.size, l.stride, l.pad, b);
- gemm_bin(m,n,k,1,a,k,b,n,c,n);
- c += n*m;
- state.input += l.c*l.h*l.w;
- }
- scale_bias(l.output, l.scales, l.batch, l.n, out_h*out_w);
- add_bias(l.output, l.biases, l.batch, l.n, out_h*out_w);
- activate_array(l.output, m*n*l.batch, l.activation);
- return;
- }
- */
-
if(l.xnor){
- binarize_filters(l.filters, l.n, l.c*l.size*l.size, l.binary_filters);
+ if (!l.align_bit_weights) {
+ binarize_weights(l.weights, l.n, l.c*l.size*l.size, l.binary_weights);
+ //printf("\n binarize_weights l.align_bit_weights = %p \n", l.align_bit_weights);
+ }
swap_binary(&l);
binarize_cpu(state.input, l.c*l.h*l.w*l.batch, l.binary_input);
state.input = l.binary_input;
@@ -444,22 +671,165 @@
int k = l.size*l.size*l.c;
int n = out_h*out_w;
- if (l.xnor && l.c%32 == 0 && AI2) {
- forward_xnor_layer(l, state);
- printf("xnor\n");
- } else {
+ float *a = l.weights;
+ float *b = state.workspace;
+ float *c = l.output;
- float *a = l.filters;
- float *b = state.workspace;
- float *c = l.output;
+ static int u = 0;
+ u++;
- for(i = 0; i < l.batch; ++i){
- im2col_cpu(state.input, l.c, l.h, l.w,
- l.size, l.stride, l.pad, b);
- gemm(0,0,m,n,k,1,a,k,b,n,1,c,n);
- c += n*m;
- state.input += l.c*l.h*l.w;
+ for(i = 0; i < l.batch; ++i){
+ //im2col_cpu(state.input, l.c, l.h, l.w, l.size, l.stride, l.pad, b);
+
+ //float *t_input = NULL;
+ //if (l.xnor) {
+ // size_t new_ldb = k + (l.lda_align - k%l.lda_align);
+ // size_t t_intput_size = new_ldb * n;
+ // t_input = calloc(t_intput_size, sizeof(float));
+ // im2col_cpu_custom_transpose(state.input, l.c, l.h, l.w, l.size, l.stride, l.pad, t_input, new_ldb);
+ //}
+ //if (l.xnor && l.size == 3 && l.stride == 1 && l.pad == 1) {}
+ //else
+ // further optimizations: im2col_bin() for XNOR, and then transpose_aling_bin()
+ im2col_cpu_custom(state.input, l.c, l.h, l.w, l.size, l.stride, l.pad, b);
+
+
+ //gemm(0,0,m,n,k,1,a,k,b,n,1,c,n);
+ //gemm_nn_custom(m, n, k, 1, a, k, b, n, c, n);
+ if (l.xnor) {
+ size_t output_size = l.outputs;
+ //float *count_output = calloc(output_size, sizeof(float));
+ //size_t bit_output_size = output_size / 8 + 1;
+ //char *bit_output = calloc(bit_output_size, sizeof(char));
+
+ size_t intput_size = n * k; // (out_h*out_w) X (l.size*l.size*l.c) : after im2col()
+ size_t bit_input_size = intput_size / 8 + 1;
+ //char *bit_input = calloc(bit_input_size, sizeof(char));
+
+ size_t weights_size = k * m; //l.size*l.size*l.c*l.n;
+ size_t bit_weights_size = weights_size / 8 + 1;
+ //char *bit_weights = calloc(bit_weights_size, sizeof(char));
+ //float *mean_arr = calloc(l.n, sizeof(float));
+
+ // test: float->bit->float
+ //get_mean_array(l.weights, weights_size, l.n, mean_arr);
+ //float_to_bit(l.weights, bit_weights, weights_size);
+ //memset(l.weights, 0, weights_size * sizeof(float));
+ //bit_to_float(bit_weights, l.weights, weights_size, l.n, mean_arr); // just for test float->bit->float
+
+ //float_to_bit(b, bit_input, intput_size);
+ //memset(b, 0, intput_size * sizeof(float));
+ //bit_to_float(bit_input, b, intput_size, 1, NULL); // just for test float->bit->float
+
+ // transpose B from NxK to KxN (x-axis (ldb = l.size*l.size*l.c) - should be multiple of 8 bits)
+ {
+ /*
+ size_t ldb_align = 256;// 8;
+
+ size_t new_ldb = k + (ldb_align - k%ldb_align); // (k / 8 + 1) * 8;
+ size_t t_intput_size = new_ldb * n;
+ size_t t_bit_input_size = t_intput_size / 8;// +1;
+ float *t_input = calloc(t_intput_size, sizeof(float));
+ char *t_bit_input = calloc(t_bit_input_size, sizeof(char));
+
+ //printf("\n bit_input_size = %d, n = %d, k = %d, ldb = %d \n", bit_input_size, n, k, n);
+ //printf("\n t_bit_input_size = %d, k = %d, n = %d, new_ldb = %d \n", t_bit_input_size, k, n, new_ldb);
+
+
+ //printf("\n align_weights_size = %d, k = %d, m = %d, lda = %d \n", align_weights_size, k, m, k);
+ //printf("\n align_bit_weights_size = %d, k = %d, m = %d, new_lda = %d \n", align_bit_weights_size, k, m, new_ldb);
+
+
+ // transpose and align B
+ int i, j;
+ for (i = 0; i < n; ++i) {
+ for (j = 0; j < k; ++j) {
+ t_input[i*new_ldb + j] = b[j*n + i];
+ }
+ }
+ float_to_bit(t_input, t_bit_input, t_intput_size);
+
+
+
+ if (!l.align_bit_weights)
+ {
+ size_t align_weights_size = new_ldb * m;
+ size_t align_bit_weights_size = align_weights_size / 8;// +1;
+ float *align_weights = calloc(align_weights_size, sizeof(float));
+ l.align_bit_weights = calloc(align_bit_weights_size, sizeof(char));
+
+ // align A without transpose
+ for (i = 0; i < m; ++i) {
+ for (j = 0; j < k; ++j) {
+ align_weights[i*new_ldb + j] = a[i*k + j];
+ }
+ }
+ float_to_bit(align_weights, l.align_bit_weights, align_weights_size);
+
+ l.mean_arr = calloc(l.n, sizeof(float));
+ get_mean_array(align_weights, align_weights_size, l.n, l.mean_arr);
+
+ free(align_weights);
+ }
+ */
+
+ /*
+ if (l.size == 3 && l.stride == 1 && l.pad == 1)
+ {
+ //binarize_weights(l.weights, l.n, l.c*l.size*l.size, l.binary_weights);
+ //printf("\n mean = %f \n", l.mean_arr[0]);
+
+ convolution_2d(l.w, l.h, l.size, l.n, l.c, l.pad, l.stride,
+ //l.weights, state.input, l.output, l.mean_arr);
+ l.binary_weights, state.input, l.output, l.mean_arr);
+ }
+ else {
+ */
+
+ //size_t ldb_align = 256; // 256 bit for AVX2
+ int ldb_align = l.lda_align;
+ size_t new_ldb = k + (ldb_align - k%ldb_align);
+ char *t_bit_input = NULL;
+ size_t t_intput_size = binary_transpose_align_input(k, n, b, &t_bit_input, ldb_align);
+ //char *t_bit_input = calloc(new_ldb * n, sizeof(char)); // for im2col_cpu_custom_transpose() only
+ //float_to_bit(t_input, t_bit_input, new_ldb * n); // for im2col_cpu_custom_transpose() only
+
+ // 5x times faster than gemm()-float32
+ // further optimizations: accelerate maxpool-layer with OpenMP/AVX
+ gemm_nn_custom_bin_mean_transposed(m, n, k, 1, l.align_bit_weights, new_ldb, t_bit_input, new_ldb, c, n, l.mean_arr);
+
+ //gemm_nn_custom_bin_mean_transposed(m, n, k, 1, bit_weights, k, t_bit_input, new_ldb, c, n, mean_arr);
+
+ //free(t_input);
+ free(t_bit_input);
+ //}
+
+ }
+
+ // for bit_input: (k * n)
+ //if (u == 8) gemm_nn_custom_bin_mean(m, n, k, 1, bit_weights, k, bit_input, n, c, n, mean_arr); // last xnor layer
+ //else gemm_nn_custom_bin_mean(m, n, k, 1, bit_weights, k, bit_input, n, c, n, NULL);
+
+ //gemm_nn_custom_bin_mean(m, n, k, 1, bit_weights, k, bit_input, n, c, n, mean_arr);
+
+ //printf("\n u = %d \n", u);
+
+ //gemm_nn_custom(m, n, k, 1, a, k, b, n, c, n);
+
+ //int j;
+ //if (u != 8) for (j = 0; j < l.n; ++j) l.biases[j] = l.biases[j] / (mean_arr[j]*2);
+
+ //free(count_output);
+ //free(bit_input);
+ //free(bit_weights);
+ //free(mean_arr);
}
+ else {
+ gemm(0, 0, m, n, k, 1, a, k, b, n, 1, c, n);
+ // bit-count to float
+ }
+ c += n*m;
+ state.input += l.c*l.h*l.w;
}
if(l.batch_normalize){
@@ -467,7 +837,9 @@
}
add_bias(l.output, l.biases, l.batch, l.n, out_h*out_w);
- activate_array(l.output, m*n*l.batch, l.activation);
+ //activate_array(l.output, m*n*l.batch, l.activation);
+ activate_array_cpu_custom(l.output, m*n*l.batch, l.activation);
+
if(l.binary || l.xnor) swap_binary(&l);
}
@@ -482,19 +854,23 @@
gradient_array(l.output, m*k*l.batch, l.activation, l.delta);
backward_bias(l.bias_updates, l.delta, l.batch, l.n, k);
+ if(l.batch_normalize){
+ backward_batchnorm_layer(l, state);
+ }
+
for(i = 0; i < l.batch; ++i){
float *a = l.delta + i*m*k;
float *b = state.workspace;
- float *c = l.filter_updates;
+ float *c = l.weight_updates;
float *im = state.input+i*l.c*l.h*l.w;
- im2col_cpu(im, l.c, l.h, l.w,
+ im2col_cpu(im, l.c, l.h, l.w,
l.size, l.stride, l.pad, b);
gemm(0,1,m,n,k,1,a,k,b,k,1,c,n);
if(state.delta){
- a = l.filters;
+ a = l.weights;
b = l.delta + i*m*k;
c = state.workspace;
@@ -511,36 +887,41 @@
axpy_cpu(l.n, learning_rate/batch, l.bias_updates, 1, l.biases, 1);
scal_cpu(l.n, momentum, l.bias_updates, 1);
- axpy_cpu(size, -decay*batch, l.filters, 1, l.filter_updates, 1);
- axpy_cpu(size, learning_rate/batch, l.filter_updates, 1, l.filters, 1);
- scal_cpu(size, momentum, l.filter_updates, 1);
+ if(l.scales){
+ axpy_cpu(l.n, learning_rate/batch, l.scale_updates, 1, l.scales, 1);
+ scal_cpu(l.n, momentum, l.scale_updates, 1);
+ }
+
+ axpy_cpu(size, -decay*batch, l.weights, 1, l.weight_updates, 1);
+ axpy_cpu(size, learning_rate/batch, l.weight_updates, 1, l.weights, 1);
+ scal_cpu(size, momentum, l.weight_updates, 1);
}
-image get_convolutional_filter(convolutional_layer l, int i)
+image get_convolutional_weight(convolutional_layer l, int i)
{
int h = l.size;
int w = l.size;
int c = l.c;
- return float_to_image(w,h,c,l.filters+i*h*w*c);
+ return float_to_image(w,h,c,l.weights+i*h*w*c);
}
-void rgbgr_filters(convolutional_layer l)
+void rgbgr_weights(convolutional_layer l)
{
int i;
for(i = 0; i < l.n; ++i){
- image im = get_convolutional_filter(l, i);
+ image im = get_convolutional_weight(l, i);
if (im.c == 3) {
rgbgr_image(im);
}
}
}
-void rescale_filters(convolutional_layer l, float scale, float trans)
+void rescale_weights(convolutional_layer l, float scale, float trans)
{
int i;
for(i = 0; i < l.n; ++i){
- image im = get_convolutional_filter(l, i);
+ image im = get_convolutional_weight(l, i);
if (im.c == 3) {
scale_image(im, scale);
float sum = sum_array(im.data, im.w*im.h*im.c);
@@ -549,21 +930,21 @@
}
}
-image *get_filters(convolutional_layer l)
+image *get_weights(convolutional_layer l)
{
- image *filters = calloc(l.n, sizeof(image));
+ image *weights = calloc(l.n, sizeof(image));
int i;
for(i = 0; i < l.n; ++i){
- filters[i] = copy_image(get_convolutional_filter(l, i));
- //normalize_image(filters[i]);
+ weights[i] = copy_image(get_convolutional_weight(l, i));
+ //normalize_image(weights[i]);
}
- return filters;
+ return weights;
}
-image *visualize_convolutional_layer(convolutional_layer l, char *window, image *prev_filters)
+image *visualize_convolutional_layer(convolutional_layer l, char *window, image *prev_weights)
{
- image *single_filters = get_filters(l);
- show_images(single_filters, l.n, window);
+ image *single_weights = get_weights(l);
+ show_images(single_weights, l.n, window);
image delta = get_convolutional_image(l);
image dc = collapse_image_layers(delta, 1);
@@ -572,6 +953,6 @@
//show_image(dc, buff);
//save_image(dc, buff);
free_image(dc);
- return single_filters;
+ return single_weights;
}
--
Gitblit v1.10.0