From 3a33d00d22ef55247fe379b8e6c53850f43a32a8 Mon Sep 17 00:00:00 2001
From: Alexey <AlexeyAB@users.noreply.github.com>
Date: Tue, 19 Jun 2018 22:29:59 +0000
Subject: [PATCH] Update Readme.md
---
src/convolutional_kernels.cu | 164 ++++++++++++++++++++++++++++++++++++++++--------------
1 files changed, 121 insertions(+), 43 deletions(-)
diff --git a/src/convolutional_kernels.cu b/src/convolutional_kernels.cu
index 9d88a88..324fc50 100644
--- a/src/convolutional_kernels.cu
+++ b/src/convolutional_kernels.cu
@@ -37,7 +37,7 @@
int i = 0;
float mean = 0;
for(i = 0; i < n; ++i){
- mean += abs(input[i*size + s]);
+ mean += fabs(input[i*size + s]);
}
mean = mean / n;
for(i = 0; i < n; ++i){
@@ -59,7 +59,7 @@
int i = 0;
float mean = 0;
for(i = 0; i < size; ++i){
- mean += abs(weights[f*size + i]);
+ mean += fabs(weights[f*size + i]);
}
mean = mean / size;
for(i = 0; i < size; ++i){
@@ -135,26 +135,24 @@
// More: http://docs.nvidia.com/deeplearning/sdk/cudnn-developer-guide/index.html#tensor_ops
const size_t input16_size = l.batch*l.c*l.w*l.h;
- static size_t max_input16_size = input16_size;
- static half* input16 = cuda_make_f16_from_f32_array(NULL, max_input16_size);
-
const size_t output16_size = l.batch*l.out_c*l.out_h*l.out_w;
- static size_t max_output16_size = output16_size;
- static half* output16 = cuda_make_f16_from_f32_array(NULL, max_output16_size);
- if (max_input16_size < input16_size) {
- max_input16_size = input16_size;
- cuda_free((float *)input16);
- input16 = cuda_make_f16_from_f32_array(state.input, max_input16_size);
+ if (*state.net.max_input16_size < input16_size) {
+ //printf("\n input16_size: cur = %zu \t max = %zu \n", input16_size, *state.net.max_input16_size);
+ *state.net.max_input16_size = input16_size;
+ if (*state.net.input16_gpu) cuda_free(*state.net.input16_gpu);
+ *state.net.input16_gpu = (float *)cuda_make_f16_from_f32_array(NULL, *state.net.max_input16_size);
}
+ float *input16 = *state.net.input16_gpu;
- if (max_output16_size < output16_size) {
- max_output16_size = output16_size;
- cuda_free((float *)output16);
- output16 = cuda_make_f16_from_f32_array(NULL, max_output16_size);
+ if (*state.net.max_output16_size < output16_size) {
+ *state.net.max_output16_size = output16_size;
+ if (*state.net.output16_gpu) cuda_free(*state.net.output16_gpu);
+ *state.net.output16_gpu = (float *)cuda_make_f16_from_f32_array(NULL, *state.net.max_output16_size);
}
+ float *output16 = *state.net.output16_gpu;
- cuda_convert_f32_to_f16(state.input, input16_size, (float *)input16);
+ cuda_convert_f32_to_f16(state.input, input16_size, input16);
//fill_ongpu(output16_size / 2, 0, (float *)output16, 1);
cudnnConvolutionForward(cudnn_handle(),
@@ -171,7 +169,51 @@
l.dstTensorDesc,
output16);
- cuda_convert_f16_to_f32((float *)output16, output16_size, l.output_gpu);
+
+ if (l.batch_normalize)
+ {
+ if (state.train) // Training
+ {
+ copy_ongpu(l.outputs*l.batch / 2, output16, 1, l.x_gpu, 1);
+ //cudaMemcpyAsync(l.x_gpu, output16, l.outputs*l.batch*sizeof(half), cudaMemcpyDefault, get_cuda_stream());
+ float one = 1;
+ float zero = 0;
+ // Batch-normalization can still take FP16 inputs and outputs, saving half the bandwidth
+ // compared to FP32, it�s just that the statistics and value adjustment should be done in FP32.
+ cudnnBatchNormalizationForwardTraining(cudnn_handle(),
+ CUDNN_BATCHNORM_SPATIAL,
+ &one,
+ &zero,
+ l.normDstTensorDescF16,
+ l.x_gpu, // input
+ l.normDstTensorDescF16,
+ output16, // output
+ l.normTensorDesc,
+ l.scales_gpu,
+ l.biases_gpu,
+ .01,
+ l.rolling_mean_gpu, // output (should be FP32)
+ l.rolling_variance_gpu, // output (should be FP32)
+ .00001,
+ l.mean_gpu, // output (should be FP32)
+ l.variance_gpu); // output (should be FP32)
+
+ cuda_convert_f16_to_f32(output16, output16_size, l.output_gpu);
+ //forward_batchnorm_layer_gpu(l, state);
+ }
+ else // Detection
+ {
+ cuda_convert_f16_to_f32(output16, output16_size, l.output_gpu);
+ normalize_gpu(l.output_gpu, l.rolling_mean_gpu, l.rolling_variance_gpu, l.batch, l.out_c, l.out_h*l.out_w);
+ scale_bias_gpu(l.output_gpu, l.scales_gpu, l.batch, l.out_c, l.out_h*l.out_w);
+ add_bias_gpu(l.output_gpu, l.biases_gpu, l.batch, l.out_c, l.out_w*l.out_h);
+ }
+ }
+ else // BIAS only
+ {
+ cuda_convert_f16_to_f32(output16, output16_size, l.output_gpu);
+ add_bias_gpu(l.output_gpu, l.biases_gpu, l.batch, l.n, l.out_w*l.out_h);
+ }
#else
@@ -188,7 +230,7 @@
&one,
l.dstTensorDesc,
l.output_gpu);
-#endif
+#endif // CUDNN_HALF
#else
@@ -205,10 +247,14 @@
}
#endif
+#ifndef CUDNN_HALF
if (l.batch_normalize) {
forward_batchnorm_layer_gpu(l, state);
- }
- add_bias_gpu(l.output_gpu, l.biases_gpu, l.batch, l.n, l.out_w*l.out_h);
+ }
+ else {
+ add_bias_gpu(l.output_gpu, l.biases_gpu, l.batch, l.n, l.out_w*l.out_h);
+ }
+#endif // no CUDNN_HALF
activate_array_ongpu(l.output_gpu, l.outputs*l.batch, l.activation);
//if(l.dot > 0) dot_error_gpu(l);
@@ -222,12 +268,13 @@
backward_bias_gpu(l.bias_updates_gpu, l.delta_gpu, l.batch, l.n, l.out_w*l.out_h);
+#ifndef CUDNN_HALF
if(l.batch_normalize){
backward_batchnorm_layer_gpu(l, state);
- //axpy_ongpu(l.outputs*l.batch, -state.net.decay, l.x_gpu, 1, l.delta_gpu, 1);
} else {
- //axpy_ongpu(l.outputs*l.batch, -state.net.decay, l.output_gpu, 1, l.delta_gpu, 1);
+ //backward_bias_gpu(l.bias_updates_gpu, l.delta_gpu, l.batch, l.n, l.out_w*l.out_h);
}
+#endif // no CUDNN_HALF
float *original_input = state.input;
if(l.xnor) state.input = l.binary_input_gpu;
@@ -238,28 +285,59 @@
#ifdef CUDNN_HALF
const size_t input16_size = l.batch*l.c*l.w*l.h;
- static size_t max_input16_size = input16_size;
- static half* input16 = cuda_make_f16_from_f32_array(NULL, max_input16_size);
-
const size_t delta16_size = l.batch*l.n*l.out_w*l.out_h;
- static size_t max_delta16_size = delta16_size;
- static half* delta16 = cuda_make_f16_from_f32_array(NULL, max_delta16_size);
-
- if (max_input16_size < input16_size) {
- max_input16_size = input16_size;
- cuda_free((float *)input16);
- input16 = cuda_make_f16_from_f32_array(state.input, max_input16_size);
- }
-
- if (max_delta16_size < delta16_size) {
- max_delta16_size = delta16_size;
- cuda_free((float *)delta16);
- delta16 = cuda_make_f16_from_f32_array(NULL, max_delta16_size);
- }
-
- cuda_convert_f32_to_f16(state.input, input16_size, (float *)input16);
- cuda_convert_f32_to_f16(l.delta_gpu, delta16_size, (float *)delta16);
+ if (*state.net.max_input16_size < input16_size) {
+ *state.net.max_input16_size = input16_size;
+ if(*state.net.input16_gpu) cuda_free(*state.net.input16_gpu);
+ *state.net.input16_gpu = (float *)cuda_make_f16_from_f32_array(NULL, *state.net.max_input16_size);
+ }
+ float *input16 = *state.net.input16_gpu;
+
+ if (*state.net.max_output16_size < delta16_size) {
+ *state.net.max_output16_size = delta16_size;
+ if(*state.net.output16_gpu) cuda_free(*state.net.output16_gpu);
+ *state.net.output16_gpu = (float *)cuda_make_f16_from_f32_array(NULL, *state.net.max_output16_size);
+ }
+ float *delta16 = *state.net.output16_gpu;
+
+ cuda_convert_f32_to_f16(state.input, input16_size, input16);
+ cuda_convert_f32_to_f16(l.delta_gpu, delta16_size, delta16);
+
+ if (l.batch_normalize) {
+ //if (!state.train) {
+ // l.mean_gpu = l.rolling_mean_gpu;
+ // l.variance_gpu = l.rolling_variance_gpu;
+ //}
+ float one = 1;
+ float zero = 0;
+ cudnnBatchNormalizationBackward(cudnn_handle(),
+ CUDNN_BATCHNORM_SPATIAL,
+ &one,
+ &zero,
+ &one,
+ &one,
+ l.normDstTensorDescF16,
+ l.x_gpu, // input
+ l.normDstTensorDescF16,
+ delta16, // input
+ l.normDstTensorDescF16,
+ l.x_norm_gpu, // output
+ l.normTensorDesc,
+ l.scales_gpu, // output (should be FP32)
+ l.scale_updates_gpu, // output (should be FP32)
+ l.bias_updates_gpu, // output (should be FP32)
+ .00001,
+ l.mean_gpu, // input (should be FP32)
+ l.variance_gpu); // input (should be FP32)
+ copy_ongpu(l.outputs*l.batch / 2, l.x_norm_gpu, 1, delta16, 1);
+ //cudaMemcpyAsync(delta16, l.x_norm_gpu, l.outputs*l.batch * sizeof(half), cudaMemcpyDefault, get_cuda_stream());
+ }
+ else
+ {
+ //backward_bias_gpu(l.bias_updates_gpu, l.delta_gpu, l.batch, l.n, l.out_w*l.out_h);
+ }
+
// convert input: state.input (x), l.delta_gpu (y) from fp32 to fp16
// get output: l.weight_updates_gpu (dw) and convert it to fp32 (ONLY if it is fp16)
@@ -305,7 +383,7 @@
l.dsrcTensorDesc,
input16); // state.delta);
- cuda_convert_f16_to_f32((float *)input16, input16_size, state.delta);
+ cuda_convert_f16_to_f32(input16, input16_size, state.delta);
if (l.binary || l.xnor) swap_binary(&l);
if (l.xnor) gradient_array_ongpu(original_input, l.batch*l.c*l.h*l.w, HARDTAN, state.delta);
--
Gitblit v1.10.0