From d9f1b0b16edeb59281355a855e18a8be343fc33c Mon Sep 17 00:00:00 2001
From: Joseph Redmon <pjreddie@gmail.com>
Date: Fri, 08 Aug 2014 19:04:15 +0000
Subject: [PATCH] probably how maxpool layers should be
---
src/mini_blas.h | 41 +++++++++++++++++++++++++++++++----------
1 files changed, 31 insertions(+), 10 deletions(-)
diff --git a/src/mini_blas.h b/src/mini_blas.h
index 56e4fa7..c80e6ad 100644
--- a/src/mini_blas.h
+++ b/src/mini_blas.h
@@ -1,3 +1,5 @@
+#include "opencl.h"
+
void pm(int M, int N, float *A);
void gemm(int TA, int TB, int M, int N, int K, float ALPHA,
float *A, int lda,
@@ -5,24 +7,43 @@
float BETA,
float *C, int ldc);
float *random_matrix(int rows, int cols);
-void im2row(float *image, int h, int w, int c, int size, int stride, float *matrix);
-void im2col(float *image, int h, int w, int c, int size, int stride, float *matrix);
-void im2col_cpu(float* data_im, const int channels,
- const int height, const int width, const int ksize, const int stride,
- float* data_col);
-void col2im_cpu(float* data_col, const int channels,
- const int height, const int width, const int ksize, const int stride,
- float* data_im);
+void time_random_matrix(int TA, int TB, int m, int k, int n);
+
+#ifdef GPU
+void im2col_ongpu(cl_mem data_im, const int batch,
+ const int channels, const int height, const int width,
+ const int ksize, const int stride, cl_mem data_col);
+
+void im2col_gpu(float *data_im,
+ const int batch, const int channels, const int height, const int width,
+ const int ksize, const int stride, float *data_col);
+
+void gemm_ongpu(int TA, int TB, int M, int N, int K, float ALPHA,
+ cl_mem A_gpu, int lda,
+ cl_mem B_gpu, int ldb,
+ float BETA,
+ cl_mem C_gpu, int ldc);
+#endif
+
+void im2col_cpu(float* data_im, const int batch,
+ const int channels, const int height, const int width,
+ const int ksize, const int stride, int pad, float* data_col);
+
+void col2im_cpu(float* data_col,
+ const int channels, const int height, const int width,
+ const int ksize, const int stride, int pad, float* data_im);
void test_blas();
-void gpu_gemm(int TA, int TB, int M, int N, int K, float ALPHA,
+void gemm_gpu(int TA, int TB, int M, int N, int K, float ALPHA,
float *A, int lda,
float *B, int ldb,
float BETA,
float *C, int ldc);
-void cpu_gemm(int TA, int TB, int M, int N, int K, float ALPHA,
+void gemm_cpu(int TA, int TB, int M, int N, int K, float ALPHA,
float *A, int lda,
float *B, int ldb,
float BETA,
float *C, int ldc);
+void axpy_cpu(int N, float ALPHA, float *X, int INCX, float *Y, int INCY);
+void scal_cpu(int N, float ALPHA, float *X, int INCX);
void test_gpu_blas();
--
Gitblit v1.10.0