From a99050f0c8cb0315fa31e3d1fa3e38594fe5e40a Mon Sep 17 00:00:00 2001
From: Joseph Redmon <pjreddie@gmail.com>
Date: Mon, 08 Dec 2014 04:16:21 +0000
Subject: [PATCH] Some fixes to momentum
---
src/softmax_layer.c | 113 +++++++++++++++++++++++++++++++++++++++++++++++++-------
1 files changed, 99 insertions(+), 14 deletions(-)
diff --git a/src/softmax_layer.c b/src/softmax_layer.c
index 28696b7..abd9abf 100644
--- a/src/softmax_layer.c
+++ b/src/softmax_layer.c
@@ -1,35 +1,120 @@
#include "softmax_layer.h"
+#include "mini_blas.h"
+#include <float.h>
#include <math.h>
#include <stdlib.h>
#include <stdio.h>
-softmax_layer *make_softmax_layer(int inputs)
+softmax_layer *make_softmax_layer(int batch, int inputs)
{
- printf("Softmax Layer: %d inputs\n", inputs);
+ fprintf(stderr, "Softmax Layer: %d inputs\n", inputs);
softmax_layer *layer = calloc(1, sizeof(softmax_layer));
+ layer->batch = batch;
layer->inputs = inputs;
- layer->output = calloc(inputs, sizeof(double));
- layer->delta = calloc(inputs, sizeof(double));
+ layer->output = calloc(inputs*batch, sizeof(float));
+ layer->delta = calloc(inputs*batch, sizeof(float));
+ layer->jacobian = calloc(inputs*inputs*batch, sizeof(float));
+ #ifdef GPU
+ layer->output_cl = cl_make_array(layer->output, inputs*batch);
+ layer->delta_cl = cl_make_array(layer->delta, inputs*batch);
+ #endif
return layer;
}
-void forward_softmax_layer(const softmax_layer layer, double *input)
+void forward_softmax_layer(const softmax_layer layer, float *input)
{
- int i;
- double sum = 0;
- for(i = 0; i < layer.inputs; ++i){
- sum += exp(input[i]);
- }
- for(i = 0; i < layer.inputs; ++i){
- layer.output[i] = exp(input[i])/sum;
+ int i,b;
+ for(b = 0; b < layer.batch; ++b){
+ float sum = 0;
+ float largest = -FLT_MAX;
+ for(i = 0; i < layer.inputs; ++i){
+ if(input[i+b*layer.inputs] > largest) largest = input[i+b*layer.inputs];
+ }
+ for(i = 0; i < layer.inputs; ++i){
+ sum += exp(input[i+b*layer.inputs]-largest);
+ }
+ if(sum) sum = largest+log(sum);
+ else sum = largest-100;
+ for(i = 0; i < layer.inputs; ++i){
+ layer.output[i+b*layer.inputs] = exp(input[i+b*layer.inputs]-sum);
+ }
}
}
-void backward_softmax_layer(const softmax_layer layer, double *input, double *delta)
+void backward_softmax_layer(const softmax_layer layer, float *delta)
{
int i;
- for(i = 0; i < layer.inputs; ++i){
+ for(i = 0; i < layer.inputs*layer.batch; ++i){
delta[i] = layer.delta[i];
}
}
+#ifdef GPU
+
+void pull_softmax_layer_output(const softmax_layer layer)
+{
+ cl_read_array(layer.output_cl, layer.output, layer.inputs*layer.batch);
+}
+
+cl_kernel get_softmax_forward_kernel()
+{
+ static int init = 0;
+ static cl_kernel kernel;
+ if(!init){
+ kernel = get_kernel("src/softmax_layer.cl", "forward", 0);
+ init = 1;
+ }
+ return kernel;
+}
+
+void forward_softmax_layer_gpu(const softmax_layer layer, cl_mem input)
+{
+ cl_setup();
+ cl_kernel kernel = get_softmax_forward_kernel();
+ cl_command_queue queue = cl.queue;
+
+ cl_uint i = 0;
+ cl.error = clSetKernelArg(kernel, i++, sizeof(layer.inputs), (void*) &layer.inputs);
+ cl.error = clSetKernelArg(kernel, i++, sizeof(input), (void*) &input);
+ cl.error = clSetKernelArg(kernel, i++, sizeof(layer.output_cl), (void*) &layer.output_cl);
+ check_error(cl);
+
+ const size_t global_size[] = {layer.batch};
+
+ cl.error = clEnqueueNDRangeKernel(queue, kernel, 1, 0, global_size, 0, 0, 0, 0);
+ check_error(cl);
+
+ /*
+ cl_read_array(layer.output_cl, layer.output, layer.inputs*layer.batch);
+ int z;
+ for(z = 0; z < layer.inputs*layer.batch; ++z) printf("%f,",layer.output[z]);
+ */
+}
+
+void backward_softmax_layer_gpu(const softmax_layer layer, cl_mem delta)
+{
+ copy_ongpu(layer.batch*layer.inputs, layer.delta_cl, 1, delta, 1);
+}
+#endif
+
+/* This is if you want softmax w/o log-loss classification. You probably don't.
+ int i,j,b;
+ for(b = 0; b < layer.batch; ++b){
+ for(i = 0; i < layer.inputs; ++i){
+ for(j = 0; j < layer.inputs; ++j){
+ int d = (i==j);
+ layer.jacobian[b*layer.inputs*layer.inputs + i*layer.inputs + j] =
+ layer.output[b*layer.inputs + i] * (d - layer.output[b*layer.inputs + j]);
+ }
+ }
+ }
+ for(b = 0; b < layer.batch; ++b){
+ int M = layer.inputs;
+ int N = 1;
+ int K = layer.inputs;
+ float *A = layer.jacobian + b*layer.inputs*layer.inputs;
+ float *B = layer.delta + b*layer.inputs;
+ float *C = delta + b*layer.inputs;
+ gemm(0,0,M,N,K,1,A,K,B,N,0,C,N);
+ }
+ */
--
Gitblit v1.10.0