static int
c_set_filter(PyGpuArrayObject *var, cudnnFilterDescriptor_t desc, size_t groups) {
cudnnDataType_t dt;
cudnnStatus_t err;
if (!GpuArray_IS_C_CONTIGUOUS(&var->ga)) {
PyErr_SetString(PyExc_ValueError,
"Only contiguous filters (kernels) are supported.");
return -1;
}
switch (var->ga.typecode) {
case GA_FLOAT:
dt = CUDNN_DATA_FLOAT;
break;
case GA_DOUBLE:
dt = CUDNN_DATA_DOUBLE;
break;
case GA_HALF:
dt = CUDNN_DATA_HALF;
break;
default:
PyErr_SetString(PyExc_TypeError, "Non-float datatype in c_set_filter");
return -1;
}
int dims[8];
unsigned int nd = PyGpuArray_NDIM(var);
if (nd > 8) {
PyErr_SetString(PyExc_TypeError, "Tensor of more than 8d");
return -1;
}
for (unsigned int _i = nd; _i > 0; _i--) {
unsigned int i = _i - 1;
dims[i] = PyGpuArray_DIM(var, i);
}
/* Filters can't be less than 3d so we pad */
for (unsigned int i = nd; i < 3; i++)
dims[i] = 1;
dims[0] = dims[0] / groups;
if (nd < 3)
nd = 3;
err = cudnnSetFilterNdDescriptor(desc, dt, CUDNN_TENSOR_NCHW, nd, dims);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"Could not set filter descriptor: %s.",
cudnnGetErrorString(err));
return -1;
}
return 0;
}
static int
c_set_tensor_for_conv(PyGpuArrayObject *var, cudnnTensorDescriptor_t desc, size_t groups) {
cudnnDataType_t dt;
size_t ds;
switch (var->ga.typecode) {
case GA_FLOAT:
dt = CUDNN_DATA_FLOAT;
break;
case GA_DOUBLE:
dt = CUDNN_DATA_DOUBLE;
break;
case GA_HALF:
dt = CUDNN_DATA_HALF;
break;
default:
PyErr_SetString(PyExc_TypeError, "Non-float datatype in c_set_tensorNd");
return -1;
}
ds = gpuarray_get_elsize(var->ga.typecode);
int strs[8], dims[8], default_stride = 1;
unsigned int nd = PyGpuArray_NDIM(var);
if (nd > 8) {
PyErr_SetString(PyExc_TypeError, "Tensor of more than 8d");
return -1;
}
for (unsigned int _i = nd; _i > 0; _i--) {
unsigned int i = _i - 1;
strs[i] = (PyGpuArray_DIM(var, i) != 1 && PyGpuArray_STRIDE(var, i)) ?
PyGpuArray_STRIDE(var, i)/ds : default_stride;
default_stride *= PyGpuArray_DIM(var, i);
dims[i] = PyGpuArray_DIM(var, i);
}
/* Tensors can't be smaller than 3d for cudnn so we pad the
* descriptor if they are */
for (unsigned int i = nd; i < 3; i++) {
strs[i] = 1;
dims[i] = 1;
}
//only for grouped convolution i.e when groups > 1
dims[1] = dims[1] / groups;
cudnnStatus_t err = cudnnSetTensorNdDescriptor(desc, dt, nd < 3 ? 3 : nd,
dims, strs);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"Could not set tensorNd descriptor: %s",
cudnnGetErrorString(err));
return -1;
}
return 0;
}
static int
c_set_filter(PyGpuArrayObject *var, cudnnFilterDescriptor_t desc) {
cudnnDataType_t dt;
if (!GpuArray_IS_C_CONTIGUOUS(&var->ga)) {
PyErr_SetString(PyExc_ValueError,
"Only contiguous filters (kernels) are supported.");
return -1;
}
switch (var->ga.typecode) {
case GA_FLOAT:
dt = CUDNN_DATA_FLOAT;
break;
case GA_DOUBLE:
dt = CUDNN_DATA_DOUBLE;
break;
#if CUDNN_VERSION > 3000
case GA_HALF:
dt = CUDNN_DATA_HALF;
break;
#endif
default:
PyErr_SetString(PyExc_TypeError, "Non-float datatype in c_set_filter");
return -1;
}
int dims[5];
unsigned int nd = PyGpuArray_NDIM(var);
if (nd > 5) {
PyErr_SetString(PyExc_TypeError, "Tensor of more than 5d");
return -1;
}
for (unsigned int _i = nd; _i > 0; _i--) {
unsigned int i = _i - 1;
dims[i] = PyGpuArray_DIM(var, i);
}
cudnnStatus_t err = cudnnSetFilterNdDescriptor(desc, dt, nd, dims);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"Could not set filter descriptor: %s.",
cudnnGetErrorString(err));
return -1;
}
return 0;
}
static int
c_set_tensorNd(PyGpuArrayObject *var, cudnnTensorDescriptor_t desc) {
cudnnDataType_t dt;
size_t ds;
switch (var->ga.typecode) {
case GA_FLOAT:
dt = CUDNN_DATA_FLOAT;
break;
case GA_DOUBLE:
dt = CUDNN_DATA_DOUBLE;
break;
#if CUDNN_VERSION > 3000
case GA_HALF:
dt = CUDNN_DATA_HALF;
break;
#endif
default:
PyErr_SetString(PyExc_TypeError, "Non-float datatype in c_set_tensorNd");
return -1;
}
ds = gpuarray_get_elsize(var->ga.typecode);
int strs[5], dims[5], default_stride = 1;
unsigned int nd = PyGpuArray_NDIM(var);
if (nd > 5) {
PyErr_SetString(PyExc_TypeError, "Tensor of more than 5d");
return -1;
}
for (unsigned int _i = nd; _i > 0; _i--) {
unsigned int i = _i - 1;
strs[i] = PyGpuArray_STRIDE(var, i) ?
PyGpuArray_STRIDE(var, i)/ds : default_stride;
default_stride *= PyGpuArray_DIM(var, i);
dims[i] = PyGpuArray_DIM(var, i);
}
cudnnStatus_t err = cudnnSetTensorNdDescriptor(desc, dt, nd, dims, strs);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"Could not set tensorNd descriptor: %s",
cudnnGetErrorString(err));
return -1;
}
return 0;
}
int
APPLY_SPECIFIC(conv_fwd)(PyGpuArrayObject *input, PyGpuArrayObject *kerns,
PyGpuArrayObject *om,
cudnnConvolutionDescriptor_t desc,
double alpha, double beta,
PyGpuArrayObject **output,
PARAMS_TYPE* params) {
PyGpuContextObject *c = input->context;
void *alpha_p;
void *beta_p;
float af = alpha, bf = beta;
cudnnStatus_t err = CUDNN_STATUS_SUCCESS;
if (PyGpuArray_DIMS(input)[1] != PyGpuArray_DIMS(kerns)[1] * params->num_groups) {
PyErr_SetString(PyExc_ValueError,
"images and kernel must have the same stack size");
return 1;
}
if ((PyGpuArray_DIMS(kerns)[0] % params->num_groups) != 0) {
PyErr_SetString(PyExc_ValueError,
"Number of filters must be divisible by number of groups");
return 1;
}
switch (input->ga.typecode) {
case GA_DOUBLE:
alpha_p = (void *)α
beta_p = (void *)β
break;
case GA_FLOAT:
case GA_HALF:
alpha_p = (void *)⁡
beta_p = (void *)&bf;
break;
default:
PyErr_SetString(PyExc_TypeError, "Unsupported type in convolution");
return 1;
}
if (params->inplace) {
Py_XDECREF(*output);
*output = om;
Py_INCREF(*output);
} else {
if (theano_prep_output(output, PyGpuArray_NDIM(om), PyGpuArray_DIMS(om),
om->ga.typecode, GA_C_ORDER, c) != 0)
return 1;
if (beta != 0.0 && pygpu_move(*output, om))
return 1;
}
if (PyGpuArray_DIMS(input)[0] == 0 || PyGpuArray_DIMS(kerns)[0] == 0 || PyGpuArray_DIMS(kerns)[1] == 0) {
int err2 = GpuArray_memset(&(*output)->ga, 0);
if (err2 != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuDnnConv could not fill the output with zeros: %d", err2);
return 1;
}
return 0;
}
if (c_set_tensor_for_conv(input, APPLY_SPECIFIC(input), params->num_groups) == -1)
return 1;
if (c_set_filter(kerns, APPLY_SPECIFIC(kerns), params->num_groups) == -1)
return 1;
if (c_set_tensor_for_conv(*output, APPLY_SPECIFIC(output), params->num_groups) == -1)
return 1;
size_t input_offset = PyGpuArray_STRIDE(input, 0) / params->num_groups;
size_t kern_offset = PyGpuArray_STRIDE(kerns, 0) * PyGpuArray_DIM(kerns, 0) / params->num_groups;
size_t output_offset = PyGpuArray_STRIDE(*output, 0) / params->num_groups;
cudnnConvolutionFwdAlgo_t algo = params->conv_algo;
#ifdef DEBUG
char algorithm_name[128];
#endif
cuda_enter(c->ctx);
if (params->choose_algo) {
if (!params->choose_once) {
reuse_algo = 1;
for (unsigned int i = 0; i < PyGpuArray_NDIM(input); i++) {
reuse_algo = (reuse_algo &&
PyGpuArray_DIM(input, i) == prev_img_dims[i]);
reuse_algo = (reuse_algo &&
PyGpuArray_DIM(kerns, i) == prev_kern_dims[i]);
}
}
if (!reuse_algo) {
size_t free;
int err2 = gpucontext_property(c->ctx, GA_CTX_PROP_LARGEST_MEMBLOCK, &free);
if (err2 != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError, "Error when trying to find the "
"memory information on the GPU");
cuda_exit(c->ctx);
return 1;
}
// Guess 4Mb if the info is not available
if (free == 0) free = 4 * 1024 * 1024;
if (params->choose_time) {
int count;
cudnnConvolutionFwdAlgoPerf_t choice;
gpudata *tmpmem;
tmpmem = gpudata_alloc(c->ctx, free, NULL, 0, NULL);
if (tmpmem == NULL) {
PyErr_SetString(PyExc_MemoryError, "Could not allocate working GPU memory");
return -1;
}
// We don't sync the buffer as we don't care about the values.
err = cudnnFindConvolutionForwardAlgorithmEx(
params->handle, APPLY_SPECIFIC(input), PyGpuArray_DEV_DATA(input),
APPLY_SPECIFIC(kerns), PyGpuArray_DEV_DATA(kerns),
desc, APPLY_SPECIFIC(output), PyGpuArray_DEV_DATA(*output),
1, &count, &choice, *(void **)tmpmem,
free);
gpudata_release(tmpmem);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error selecting convolution algo: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
algo = choice.algo;
#ifdef DEBUG
if (count == 0) {
PyErr_SetString(PyExc_RuntimeError, "No best-timed conv fwd algorithm found");
return 1;
} else if (choice.status != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error getting best-timed FWD algo: %s",
cudnnGetErrorString(choice.status));
return 1;
} // Else, count is necessarly 1 for current implementation.
#endif
} else {
err = cudnnGetConvolutionForwardAlgorithm(
params->handle, APPLY_SPECIFIC(input), APPLY_SPECIFIC(kerns),
desc, APPLY_SPECIFIC(output),
CUDNN_CONVOLUTION_FWD_SPECIFY_WORKSPACE_LIMIT, free, &algo);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error selecting convolution algo: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
}
prev_algo = algo;
} else {
algo = prev_algo;
}
#ifdef DEBUG
if (0 != theano_enum_to_string_cudnnConvolutionFwdAlgo_t(algo, algorithm_name))
return 1;
// NB: This is printed only when algorithm is chosen at runtime.
if (reuse_algo)
fprintf(stderr, "(reused %s)\n", algorithm_name);
else
fprintf(stderr, "(using %s)\n", algorithm_name);
#endif
if (params->choose_once) {
reuse_algo = 1;
} else {
for (unsigned int i = 0; i < PyGpuArray_NDIM(input); i++) {
prev_img_dims[i] = PyGpuArray_DIM(input, i);
prev_kern_dims[i] = PyGpuArray_DIM(kerns, i);
}
}
}
/* Only these algos are supported for 3d conv with cuDNN >= V5.1. */
if (PyGpuArray_NDIM(input) == 5 &&
!(algo == CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM ||
algo == CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM ||
algo == CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING))
{
#ifdef DEBUG
if (0 != theano_enum_to_string_cudnnConvolutionFwdAlgo_t(algo, algorithm_name))
return 1;
fprintf(stderr, "(%s unsupported for 3D: fallback to CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM)\n", algorithm_name);
#endif
algo = CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM;
}
// Algo `small` does not work for a batch size > 2^16, with cuDNN >= V5.1.
// Issue should be resolved for cuDNN > V6.0.
if (cudnnGetVersion() < 6100 &&
algo == CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM &&
PyGpuArray_DIM(input, 0) > 65536)
{
#ifdef DEBUG
fprintf(stderr, "(CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM "
"will fail with batch size > 2^16, fallback to CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM)\n");
#endif
algo = CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM;
}
// The FFT implementation does not support strides, 1x1 filters or inputs
// with a spatial dimension larger than 1024. The tiled-FFT implementation
// does not support strides.
// If the chosen implementation is FFT or tiled-FFT, validate that it can
// be used on the current data and default to a safe implementation if it
// can't.
// The following code is 2d-specific but it is fine as FFT and tiled-FFT are
// defined only for 2d filters
/* NB:
TODO: These checkings seems outdated for FFT algorithms with cuDNN >= 5.1.
New conditions apply and may depend on number of dimensions (2D or 3D)
e.g. for FFT_TILING.
TODO: More globally, how to handle CUDNN_STATUS_NOT_SUPPORTED with unsupported algorithms?
*/
if ((algo == CUDNN_CONVOLUTION_FWD_ALGO_FFT ||
algo == CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING) && PyGpuArray_NDIM(input) == 4) {
// Extract the properties of the convolution descriptor
int nd;
int pad[2];
int stride[2];
int dilation[2];
cudnnConvolutionMode_t mode;
cudnnDataType_t data_type;
err = cudnnGetConvolutionNdDescriptor(desc, 2, &nd, pad, stride,
dilation, &mode, &data_type);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error getting convolution properties: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
if (algo == CUDNN_CONVOLUTION_FWD_ALGO_FFT) {
if (stride[0] != 1 || stride[1] != 1 ||
PyGpuArray_DIM(input, 2) > 1024 || PyGpuArray_DIM(input, 3) > 1024 ||
(PyGpuArray_DIM(kerns, 2) == 1 && PyGpuArray_DIM(kerns, 3) == 1))
{
algo = CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM;
}
} else {
// algo == CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING
if (stride[0] != 1 || stride[1] != 1) {
algo = CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM;
}
}
}
{
size_t worksize;
gpudata *workspace;
err = cudnnGetConvolutionForwardWorkspaceSize(params->handle,
APPLY_SPECIFIC(input),
APPLY_SPECIFIC(kerns),
desc,
APPLY_SPECIFIC(output),
algo,
&worksize);
if (err == CUDNN_STATUS_NOT_SUPPORTED) {
// Fallback to none algo if not supported
#ifdef DEBUG
if (0 != theano_enum_to_string_cudnnConvolutionFwdAlgo_t(algo, algorithm_name))
return 1;
fprintf(stderr, "(%s error getting worksize: "
"fallback to CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM)\n", algorithm_name);
#endif
algo = CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM;
err = cudnnGetConvolutionForwardWorkspaceSize(params->handle,
APPLY_SPECIFIC(input),
APPLY_SPECIFIC(kerns),
desc,
APPLY_SPECIFIC(output),
algo,
&worksize);
}
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error getting worksize: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
/*
* This is less than ideal since we need to free it after (which
* introduces a synchronization point. But we don't have a module
* to place a nice get_work_mem() function in.
*/
if (worksize != 0) {
workspace = gpudata_alloc(c->ctx, worksize, NULL, 0, NULL);
if (workspace == NULL) {
PyErr_SetString(PyExc_RuntimeError,
"Could not allocate working memory");
cuda_exit(c->ctx);
return 1;
}
}
cuda_wait(input->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_wait(kerns->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_wait((*output)->ga.data, GPUARRAY_CUDA_WAIT_WRITE);
for ( int g = 0; g < params->num_groups; g++) {
err = cudnnConvolutionForward(
params->handle,
alpha_p,
APPLY_SPECIFIC(input), ((char *)PyGpuArray_DEV_DATA(input)) + input_offset * g,
APPLY_SPECIFIC(kerns), ((char *)PyGpuArray_DEV_DATA(kerns)) + kern_offset * g,
desc, algo,
worksize == 0 ? NULL : *(void **)workspace, worksize,
beta_p,
APPLY_SPECIFIC(output), ((char *)PyGpuArray_DEV_DATA(*output)) + output_offset * g);
}
if (worksize != 0)
gpudata_release(workspace);
cuda_record(input->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_record(kerns->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_record((*output)->ga.data, GPUARRAY_CUDA_WAIT_WRITE);
}
cuda_exit(c->ctx);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError, "error doing operation: %s",
cudnnGetErrorString(err));
return 1;
}
return 0;
}
// Theano op code
// Authors: Arjun Jain, Frederic Bastien, Jan Schluter
// Reference code: https://github.com/BVLC/caffe/blob/master/src/caffe/layers/conv_layer.cu
// and https://github.com/torch/cunn/blob/master/SpatialConvolutionMM.cu
// Adaptation for 3d
PyGpuArrayObject* corr3dMM(PyGpuArrayObject *const bottom,
PyGpuArrayObject *const weight,
PyGpuArrayObject *const top,
const size_t direction,
const size_t dH = 1,
const size_t dW = 1,
const size_t dD = 1,
const size_t dilH = 1,
const size_t dilW = 1,
const size_t dilD = 1,
const size_t padH = 0,
const size_t padW = 0,
const size_t padD = 0)
{
if (PyGpuArray_NDIM(bottom) != 5)
{
PyErr_SetString(PyExc_ValueError, "GpuCorr3dMM requires bottom of 5D");
return NULL;
}
if (!GpuArray_IS_C_CONTIGUOUS(&bottom->ga))
{
PyErr_Format(PyExc_ValueError,
"GpuCorr3dMM requires bottom to be C-contiguous, "
"but strides are: %ld %ld %ld %ld %ld\n",
PyGpuArray_STRIDES(bottom)[0],
PyGpuArray_STRIDES(bottom)[1],
PyGpuArray_STRIDES(bottom)[2],
PyGpuArray_STRIDES(bottom)[3],
PyGpuArray_STRIDES(bottom)[4]);
return NULL;
}
if (PyGpuArray_NDIM(weight) != 5)
{
PyErr_SetString(PyExc_ValueError, "GpuCorr3dMM requires weight of 5D");
return NULL;
}
if (!GpuArray_IS_C_CONTIGUOUS(&weight->ga))
{
PyErr_Format(PyExc_ValueError,
"GpuCorr3dMM requires weight to be C-contiguous, "
"but strides are: %ld %ld %ld %ld %ld\n",
PyGpuArray_STRIDES(weight)[0],
PyGpuArray_STRIDES(weight)[1],
PyGpuArray_STRIDES(weight)[2],
PyGpuArray_STRIDES(weight)[3],
PyGpuArray_STRIDES(weight)[4]);
return NULL;
}
if (PyGpuArray_NDIM(top) != 5)
{
PyErr_SetString(PyExc_ValueError, "GpuCorr3dMM requires top of 5D");
return NULL;
}
if (!GpuArray_IS_C_CONTIGUOUS(&top->ga))
{
PyErr_Format(PyExc_ValueError,
"GpuCorr3dMM requires top to be C-contiguous, "
"but strides are: %ld %ld %ld %ld %ld\n",
PyGpuArray_STRIDES(top)[0],
PyGpuArray_STRIDES(top)[1],
PyGpuArray_STRIDES(top)[2],
PyGpuArray_STRIDES(top)[3],
PyGpuArray_STRIDES(top)[4]);
return NULL;
}
// Extract some shape information for later and check shape consistency
// bottom: (batchSize, nChannels, bottomHeight, bottomWidth, bottomDepth)
const size_t batchSize = PyGpuArray_DIMS(bottom)[0];
const size_t nChannels = PyGpuArray_DIMS(bottom)[1];
const size_t bottomHeight = PyGpuArray_DIMS(bottom)[2];
const size_t bottomWidth = PyGpuArray_DIMS(bottom)[3];
const size_t bottomDepth = PyGpuArray_DIMS(bottom)[4];
// weights: (nFilters, nChannels, rows, columns, slices)
const size_t nFilters = PyGpuArray_DIMS(weight)[0];
const size_t kH = PyGpuArray_DIMS(weight)[2];
const size_t kW = PyGpuArray_DIMS(weight)[3];
const size_t kD = PyGpuArray_DIMS(weight)[4];
if (nChannels != PyGpuArray_DIMS(weight)[1]) {
PyErr_SetString(PyExc_ValueError,
"GpuCorr3dMM images and kernel must have the same stack size\n");
return NULL;
}
// implicit dilated filter
const size_t dil_kH = (kH - 1) * dilH + 1;
const size_t dil_kW = (kW - 1) * dilW + 1;
const size_t dil_kD = (kD - 1) * dilD + 1;
// top: (batchSize, nFilters, topHeight, topWidth, topDepth)
const size_t topHeight = (bottomHeight + 2*padH - dil_kH) / dH + 1;
const size_t topWidth = (bottomWidth + 2*padW - dil_kW) / dW + 1;
const size_t topDepth = (bottomDepth + 2*padD - dil_kD) / dD + 1;
if (batchSize != PyGpuArray_DIMS(top)[0] ||
nFilters != PyGpuArray_DIMS(top)[1] ||
topHeight != PyGpuArray_DIMS(top)[2] ||
topWidth != PyGpuArray_DIMS(top)[3] ||
topDepth != PyGpuArray_DIMS(top)[4]) {
PyErr_Format(PyExc_ValueError,
"GpuCorr3dMM shape inconsistency:\n"
" bottom shape: %ld %ld %ld %ld %ld\n"
" weight shape: %ld %ld %ld %ld %ld\n"
" top shape: %ld %ld %ld %ld %ld (expected %ld %ld %ld %ld %ld)\n",
batchSize, nChannels, bottomHeight, bottomWidth, bottomDepth,
nFilters, nChannels, kH, kW, kD,
PyGpuArray_DIMS(top)[0], PyGpuArray_DIMS(top)[1],
PyGpuArray_DIMS(top)[2], PyGpuArray_DIMS(top)[3], PyGpuArray_DIMS(top)[4],
batchSize, nFilters, topHeight, topWidth, topDepth);
return NULL;
}
int err = gpublas_setup(bottom->context->ctx);
if (err != GA_NO_ERROR) {
PyErr_SetString(PyExc_RuntimeError, "Can't setup blas");
return NULL;
}
// Get the max threads per blocks
size_t max_threads_dim;
err = gpucontext_property(bottom->context->ctx, GA_CTX_PROP_MAXLSIZE, &max_threads_dim);
if (err != GA_NO_ERROR){
PyErr_Format(PyExc_RuntimeError,
"Could not fetch max_threads_dim.");
return NULL;
}
// Create temporary columns
size_t col_dim[2];
col_dim[0] = nChannels * kW * kH * kD;
col_dim[1] = topHeight * topWidth * topDepth;
PyGpuArrayObject* col = (PyGpuArrayObject*)pygpu_empty(2, col_dim,
bottom->ga.typecode,
GA_C_ORDER,
bottom->context,
Py_None);
if (NULL == col)
{
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM failed to allocate working memory of %ld x %ld\n",
col_dim[0], col_dim[1]);
return NULL;
}
// Define some useful variables
const size_t bottom_stride = PyGpuArray_STRIDES(bottom)[0] / gpuarray_get_elsize(bottom->ga.typecode);
const size_t top_stride = PyGpuArray_STRIDES(top)[0] / gpuarray_get_elsize(top->ga.typecode);
const size_t K_ = col_dim[0];
const size_t N_ = col_dim[1];
const size_t M_ = nFilters;
const DTYPE_INPUT_0 one = 1.0f;
const DTYPE_INPUT_0 zero = 0.0f;
PyGpuArrayObject *output;
if (direction == 0) { // forward pass
output = top;
// valid correlation: im3d2col, then gemm
// Iterate over batch
for (size_t n = 0; n < batchSize; n++) {
// First, im3d2col
err = im3d2col(max_threads_dim,
bottom->ga.data, n * bottom_stride, nChannels, bottomHeight,
bottomWidth, bottomDepth, kH, kW, kD, dilH, dilW, dilD,
padH, padW, padD, dH, dW, dD, col->ga.data);
if (err != GA_NO_ERROR) {
Py_DECREF(col);
return NULL;
}
// Second, gemm
err = gpublas_sgemm(cb_fortran, cb_no_trans, cb_no_trans,
N_, M_, K_, one,
col->ga.data, 0, N_,
weight->ga.data, 0, K_,
zero,
top->ga.data, n * top_stride, N_);
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM encountered an error running sgemm.\n");
Py_DECREF(col);
return NULL;
}
}
}
else if (direction == 1) { // backprop wrt. weights
output = weight;
// valid convolution: im3col, then gemm
// Iterate over batch
for (size_t n = 0; n < batchSize; n++) {
// First, im3d2col
err = im3d2col(max_threads_dim,
bottom->ga.data, n * bottom_stride, nChannels, bottomHeight,
bottomWidth, bottomDepth, kH, kW, kD, dilH, dilW, dilD,
padH, padW, padD, dH, dW, dD, col->ga.data);
if (err != GA_NO_ERROR) {
Py_DECREF(col);
return NULL;
}
// Second, gemm
// Note that we accumulate into weight. We do so by setting beta = 0
// for the first iteration and beta = 1 for subsequent ones. (This
// is faster than setting weight to all zeros before the loop.)
err = gpublas_sgemm(cb_fortran, cb_trans, cb_no_trans,
K_, M_, N_, one,
col->ga.data, 0, N_,
top->ga.data, n * top_stride, N_,
(n == 0) ? zero : one,
weight->ga.data, 0, K_);
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM encountered an error running sgemm.\n");
Py_DECREF(col);
return NULL;
}
}
}
else if (direction == 2) { // backprop wrt. inputs
output = bottom;
// full convolution: gemm, then col2im3d
// Iterate over batch
for (size_t n = 0; n < batchSize; n++) {
// gemm into columns
err = gpublas_sgemm(cb_fortran, cb_no_trans, cb_trans,
N_, K_, M_, one,
top->ga.data, n * top_stride, N_,
weight->ga.data, 0, K_,
zero,
col->ga.data, 0, N_);
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM encountered an error running sgemm.\n");
Py_DECREF(col);
return NULL;
}
// col2im3d back to the data
err = col2im3d(max_threads_dim,
col->ga.data, nChannels,
bottomHeight, bottomWidth, bottomDepth,
kH, kW, kD, dilH, dilW, dilD, padH, padW, padD,
dH, dW, dD, bottom->ga.data, n * bottom_stride);
if (err != GA_NO_ERROR) {
Py_DECREF(col);
return NULL;
}
}
}
// Free temporary columns
Py_DECREF(col);
// Note that we don't change the refcount of the output matrix here. Output
// (re)allocation and refcounting is done in BaseGpuCorr3dMM.c_code_helper();
// in here output is just aliased to one of bottom, weights, or top.
return output;
}
int APPLY_SPECIFIC(ave_pool_grad)(PyGpuArrayObject *x,
PyGpuArrayObject *gz,
PyArrayObject *ws,
PyArrayObject *stride,
PyArrayObject *pad,
PyGpuArrayObject **gx,
PyGpuContextObject *ctx) {
if (!GpuArray_IS_C_CONTIGUOUS(&x->ga)
|| !GpuArray_IS_C_CONTIGUOUS(&gz->ga))
{
PyErr_Format(PyExc_ValueError,
"GpuMaxPoolGrad: requires data to be C-contiguous");
return 1;
}
size_t ndims = PyArray_DIM(ws, 0);
if (PyGpuArray_NDIM(x) != ndims + 2
|| PyGpuArray_NDIM(gz) != ndims + 2)
{
PyErr_SetString(PyExc_ValueError, "GpuMaxPoolGrad: rank error");
return 1;
}
if (theano_prep_output(gx, PyGpuArray_NDIM(x), PyGpuArray_DIMS(x),
x->ga.typecode, GA_C_ORDER, ctx) != 0)
{
PyErr_SetString(PyExc_RuntimeError,
"GpuMaxPoolGrad: failed to allocate memory");
return 1;
}
{
// scope for running kernel
size_t w[3];
size_t s[3];
size_t p[3];
for(int i = 0; i < ndims; i++) {
w[i] = *((npy_intp*)PyArray_GETPTR1(ws, i));
s[i] = *((npy_intp*)PyArray_GETPTR1(stride, i));
p[i] = *((npy_intp*)PyArray_GETPTR1(pad, i));
}
int err;
const size_t* z_dims = PyGpuArray_DIMS(gz);
const size_t* x_dims = PyGpuArray_DIMS(x);
if (ndims == 2) {
size_t num_kernels = x_dims[0] * x_dims[1] * x_dims[2] * x_dims[3];
err = ave_pool2d_grad_kernel_scall(1, &num_kernels, 0, num_kernels,
x_dims[0], x_dims[1], x_dims[2], x_dims[3],
z_dims[2], z_dims[3],
x->ga.data, gz->ga.data,
w[0], w[1], s[0], s[1], p[0], p[1],
INC_PAD, SUM_MODE, (*gx)->ga.data);
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuAveragePoolGrad: ave_pool2d_grad_kernel %s.",
GpuKernel_error(&k_ave_pool2d_grad_kernel, err));
return 1;
}
} else if (ndims == 3) {
size_t num_kernels = x_dims[0] * x_dims[1] * x_dims[2] * x_dims[3] * x_dims[4];
err = ave_pool3d_grad_kernel_scall(1, &num_kernels, 0, num_kernels,
x_dims[0], x_dims[1], x_dims[2], x_dims[3], x_dims[4],
z_dims[2], z_dims[3], z_dims[4],
x->ga.data, gz->ga.data,
w[0], w[1], w[2], s[0], s[1], s[2],
p[0], p[1], p[2], INC_PAD, SUM_MODE,
(*gx)->ga.data);
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuAveragePoolGrad: ave_pool3d_grad_kernel %s.",
GpuKernel_error(&k_ave_pool3d_grad_kernel, err));
return 1;
}
}
}
return 0;
}
int
APPLY_SPECIFIC(conv_gw)(PyGpuArrayObject *input, PyGpuArrayObject *output,
PyGpuArrayObject *km,
cudnnConvolutionDescriptor_t desc,
double alpha, double beta, PyGpuArrayObject **kerns,
PyGpuContextObject *c) {
cudnnStatus_t err = CUDNN_STATUS_SUCCESS;
float af = alpha, bf = beta;
void *alpha_p;
void *beta_p;
if (PyGpuArray_DIMS(input)[1] != PyGpuArray_DIMS(km)[1]) {
PyErr_SetString(PyExc_ValueError,
"GpuDnnConv images and kernel must have the same stack size");
return 1;
}
if (c_set_tensorNd(input, APPLY_SPECIFIC(input)) == -1)
return 1;
if (c_set_tensorNd(output, APPLY_SPECIFIC(output)) == -1)
return 1;
switch (input->ga.typecode) {
case GA_DOUBLE:
alpha_p = (void *)α
beta_p = (void *)β
break;
case GA_FLOAT:
case GA_HALF:
alpha_p = (void *)⁡
beta_p = (void *)&bf;
break;
default:
PyErr_SetString(PyExc_TypeError, "Unsupported type in convolution");
return 1;
}
#ifdef CONV_INPLACE
Py_XDECREF(*kerns);
*kerns = km;
Py_INCREF(*kerns);
#else
if (theano_prep_output(kerns, PyGpuArray_NDIM(km), PyGpuArray_DIMS(km),
km->ga.typecode, GA_C_ORDER, c) != 0)
return 1;
if (beta != 0.0 && pygpu_move(*kerns, km))
return 1;
#endif
if (c_set_filter(*kerns, APPLY_SPECIFIC(kerns)) == -1)
return 1;
cudnnConvolutionBwdFilterAlgo_t algo = CONV_ALGO;
cuda_enter(c->ctx);
#ifdef CHOOSE_ALGO
static int reuse_algo = 0;
static cudnnConvolutionBwdFilterAlgo_t prev_algo = CONV_ALGO;
#ifndef CHOOSE_ONCE
static size_t prev_img_dims[5] = {0};
static size_t prev_top_dims[5] = {0};
reuse_algo = 1;
for (unsigned int i = 0; i < PyGpuArray_NDIM(input); i++) {
reuse_algo = (reuse_algo &&
PyGpuArray_DIM(input, i) == prev_img_dims[i]);
reuse_algo = (reuse_algo &&
PyGpuArray_DIM(output, i) == prev_top_dims[i]);
}
#endif
if (!reuse_algo) {
#ifdef CHOOSE_TIME
int count;
cudnnConvolutionBwdFilterAlgoPerf_t choice;
err = cudnnFindConvolutionBackwardFilterAlgorithm(
APPLY_SPECIFIC(_handle), APPLY_SPECIFIC(input), APPLY_SPECIFIC(output), desc,
APPLY_SPECIFIC(kerns), 1, &count, &choice);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error selecting convolution algo: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
algo = choice.algo;
#else
size_t free = 0, total = 0;
cudaError_t err2 = cudaMemGetInfo(&free, &total);
if (err2 != cudaSuccess){
cudaGetLastError();
PyErr_Format(PyExc_RuntimeError, "Error when trying to find the memory "
"information on the GPU: %s\n", cudaGetErrorString(err2));
cuda_exit(c->ctx);
return 1;
}
err = cudnnGetConvolutionBackwardFilterAlgorithm(
APPLY_SPECIFIC(_handle), APPLY_SPECIFIC(input), APPLY_SPECIFIC(output),
desc, APPLY_SPECIFIC(kerns),
CUDNN_CONVOLUTION_BWD_FILTER_SPECIFY_WORKSPACE_LIMIT, free, &algo);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error selecting convolution algo: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
#endif
prev_algo = algo;
} else {
algo = prev_algo;
}
#ifdef CHOOSE_ONCE
reuse_algo = 1;
#else
for (unsigned int i = 0; i < PyGpuArray_NDIM(input); i++) {
prev_img_dims[i] = PyGpuArray_DIM(input, i);
prev_top_dims[i] = PyGpuArray_DIM(output, i);
}
#endif
#endif
// The FFT implementation does not support strides, 1x1 filters or inputs
// with a spatial dimension larger than 1024.
// If the chosen implementation is FFT, validate that it can
// be used on the current data and default to a safe implementation if it
// can't.
// The following code is 2d-specific but it is fine as FFT and tiled-FFT are
// defined only for 2d filters
if (algo == CUDNN_CONVOLUTION_BWD_FILTER_ALGO_FFT &&
PyGpuArray_NDIM(input) == 4) {
// Extract the properties of the convolution descriptor
int nd;
int pad[2];
int stride[2];
int upscale[2];
cudnnConvolutionMode_t mode;
cudnnDataType_t data_type;
err = cudnnGetConvolutionNdDescriptor_v3(desc, 2, &nd, pad, stride,
upscale, &mode, &data_type);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error getting convolution properties: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
if (stride[0] != 1 || stride[1] != 1 ||
PyGpuArray_DIM(input, 2) > 1024 || PyGpuArray_DIM(input, 3) > 1024 ||
(PyGpuArray_DIM(*kerns, 2) == 1 && PyGpuArray_DIM(*kerns, 3) == 1)) {
algo = CUDNN_CONVOLUTION_BWD_FILTER_ALGO_0;
}
}
size_t worksize;
gpudata *workspace;
err = cudnnGetConvolutionBackwardFilterWorkspaceSize(
APPLY_SPECIFIC(_handle), APPLY_SPECIFIC(input), APPLY_SPECIFIC(output), desc,
APPLY_SPECIFIC(kerns), algo, &worksize);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError, "error getting worksize: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
if (worksize != 0) {
workspace = c->ops->buffer_alloc(c->ctx, worksize, NULL, 0, NULL);
if (workspace == NULL) {
PyErr_SetString(PyExc_RuntimeError, "Could not allocate working memory");
cuda_exit(c->ctx);
return 1;
}
}
cuda_wait(input->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_wait(output->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_wait((*kerns)->ga.data, GPUARRAY_CUDA_WAIT_WRITE);
err = cudnnConvolutionBackwardFilter_v3(
APPLY_SPECIFIC(_handle),
alpha_p,
APPLY_SPECIFIC(input), PyGpuArray_DEV_DATA(input),
APPLY_SPECIFIC(output), PyGpuArray_DEV_DATA(output),
desc, algo, worksize == 0 ? NULL : *(void **)workspace, worksize,
beta_p,
APPLY_SPECIFIC(kerns), PyGpuArray_DEV_DATA(*kerns));
if (worksize != 0)
c->ops->buffer_release(workspace);
cuda_record(input->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_record(output->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_record((*kerns)->ga.data, GPUARRAY_CUDA_WAIT_WRITE);
cuda_exit(c->ctx);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError, "error doing operation: %s",
cudnnGetErrorString(err));
return 1;
}
return 0;
}
int APPLY_SPECIFIC(magma_svd)(PyGpuArrayObject *A,
PyGpuArrayObject **S,
PyGpuArrayObject **U, // may be NULL
PyGpuArrayObject **VT, // may be NULL
PARAMS_TYPE* params) {
bool compute_uv = (U != NULL);
magma_int_t *iwork = NULL, iunused[1];
magma_int_t M, N, K, ldu, ldv, M_U, N_VT, info;
magma_vec_t jobz;
size_t s_dims[1], u_dims[2], vt_dims[2];
float *a_data = NULL, *s_data = NULL, *u_data = NULL, *vt_data = NULL,
*work = NULL;
float dummy[1];
int res = -1, lwork;
if (A->ga.typecode != GA_FLOAT) {
PyErr_SetString(PyExc_TypeError,
"GpuMagmaMatrixInverse: Unsupported data type");
return -1;
}
// This is early to match the exit() in the fail label.
cuda_enter(params->context->ctx);
magma_init();
if (!GpuArray_IS_C_CONTIGUOUS(&A->ga)) {
PyErr_SetString(PyExc_ValueError,
"GpuMagmaMatrixInverse: requires data to be C-contiguous");
return 1;
}
if (PyGpuArray_NDIM(A) != 2) {
PyErr_SetString(PyExc_ValueError,
"GpuMagmaMatrixInverse: matrix rank error");
goto fail;
}
// magma matrix svd
// reverse dimensions because MAGMA expects column-major matrices:
M = PyGpuArray_DIM(A, 1);
N = PyGpuArray_DIM(A, 0);
K = std::min(M, N);
if (MAGMA_SUCCESS != magma_smalloc_pinned(&a_data, M * N)) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaSVD: failed to allocate memory");
goto fail;
}
cudaMemcpy(a_data, PyGpuArray_DEV_DATA(A), M * N * sizeof(float),
cudaMemcpyDeviceToDevice);
if (MAGMA_SUCCESS != magma_smalloc_pinned(&s_data, K)) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaSVD: failed to allocate memory");
goto fail;
}
if (compute_uv) {
if (params->full_matrices) {
jobz = MagmaAllVec;
} else {
jobz = MagmaSomeVec;
}
M_U = (jobz == MagmaAllVec ? M : K);
N_VT = (jobz == MagmaAllVec ? N : K);
ldu = M;
ldv = N_VT;
if (MAGMA_SUCCESS != magma_smalloc_pinned(&u_data, M_U * M)) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaSVD: failed to allocate memory");
goto fail;
}
if (MAGMA_SUCCESS != magma_smalloc_pinned(&vt_data, N * N_VT)) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaSVD: failed to allocate memory");
goto fail;
}
} else {
jobz = MagmaNoVec;
ldu = M;
ldv = N;
}
// query for workspace size
magma_sgesdd(jobz, M, N, NULL, M, NULL, NULL, ldu, NULL, ldv,
dummy, -1, iunused, &info);
lwork = (magma_int_t) MAGMA_S_REAL(dummy[0]);
if (MAGMA_SUCCESS != magma_smalloc_pinned(&work, lwork)) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaSVD: failed to allocate working memory");
goto fail;
}
if (MAGMA_SUCCESS != magma_imalloc_cpu(&iwork, 8*K)) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaSVD: failed to allocate working memory");
goto fail;
}
// compute svd
magma_sgesdd(jobz, M, N, a_data, M, s_data,
u_data, ldu, vt_data, ldv, work, lwork, iwork, &info);
if (info > 0) {
PyErr_Format(
PyExc_RuntimeError,
"GpuMagmaSVD: the updating process of SBDSDC did not converge (error: %d)",
info);
goto fail;
} else if (info < 0) {
PyErr_Format(
PyExc_RuntimeError,
"GpuMagmaSVD: magma_sgesdd_gpu argument %d has an illegal value", -info);
goto fail;
}
s_dims[0] = K;
if (theano_prep_output(S, 1, s_dims, A->ga.typecode, GA_C_ORDER, params->context) != 0){
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaSVD: failed to allocate memory");
goto fail;
}
cudaMemcpy(PyGpuArray_DEV_DATA(*S), s_data, K * sizeof(float),
cudaMemcpyDeviceToDevice);
if (compute_uv) {
u_dims[0] = N; u_dims[1] = N_VT;
if (theano_prep_output(U, 2, u_dims, A->ga.typecode, GA_C_ORDER, params->context) != 0){
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaSVD: failed to allocate memory");
goto fail;
}
// magma expects column-major matrices. Exchange u_data -> VT and vt_data -> U
// to match numpy.linalg.svd output
cudaMemcpy(PyGpuArray_DEV_DATA(*U), vt_data, N * N_VT * sizeof(float),
cudaMemcpyDeviceToDevice);
vt_dims[0] = M_U; vt_dims[1] = M;
if (theano_prep_output(VT, 2, vt_dims, A->ga.typecode, GA_C_ORDER, params->context) != 0){
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaSVD: failed to allocate memory");
goto fail;
}
// magma expects column-major matrices. Exchange u_data -> VT and vt_data -> U
// to match numpy.linalg.svd output
cudaMemcpy(PyGpuArray_DEV_DATA(*VT), u_data, M_U * M * sizeof(float),
cudaMemcpyDeviceToDevice);
}
res = 0;
fail:
if (a_data != NULL)
magma_free_pinned(a_data);
if (s_data != NULL)
magma_free_pinned(s_data);
if (u_data != NULL)
magma_free_pinned(u_data);
if (vt_data != NULL)
magma_free_pinned(vt_data);
if (work != NULL)
magma_free_pinned(work);
if (iwork != NULL)
magma_free_cpu(iwork);
magma_finalize();
cuda_exit(params->context->ctx);
return res;
}
int
APPLY_SPECIFIC(conv_fwd)(PyGpuArrayObject *input, PyGpuArrayObject *kerns,
PyGpuArrayObject *om,
cudnnConvolutionDescriptor_t desc,
double alpha, double beta,
PyGpuArrayObject **output,
PyGpuContextObject *c) {
cudnnStatus_t err = CUDNN_STATUS_SUCCESS;
float af = alpha, bf = beta;
void *alpha_p;
void *beta_p;
if (PyGpuArray_DIMS(input)[1] != PyGpuArray_DIMS(kerns)[1]) {
PyErr_SetString(PyExc_ValueError,
"images and kernel must have the same stack size");
return 1;
}
if (c_set_tensorNd(input, APPLY_SPECIFIC(input)) == -1)
return 1;
if (c_set_filter(kerns, APPLY_SPECIFIC(kerns)) == -1)
return 1;
switch (input->ga.typecode) {
case GA_DOUBLE:
alpha_p = (void *)α
beta_p = (void *)β
break;
case GA_FLOAT:
case GA_HALF:
alpha_p = (void *)⁡
beta_p = (void *)&bf;
break;
default:
PyErr_SetString(PyExc_TypeError, "Unsupported type in convolution");
return 1;
}
#ifdef CONV_INPLACE
Py_XDECREF(*output);
*output = om;
Py_INCREF(*output);
#else
if (theano_prep_output(output, PyGpuArray_NDIM(om), PyGpuArray_DIMS(om),
om->ga.typecode, GA_C_ORDER, c) != 0)
return 1;
if (beta != 0.0 && pygpu_move(*output, om))
return 1;
#endif
if (c_set_tensorNd(*output, APPLY_SPECIFIC(output)) == -1)
return 1;
cudnnConvolutionFwdAlgo_t algo = CONV_ALGO;
cuda_enter(c->ctx);
#ifdef CHOOSE_ALGO
/* Static variables are only initialized once so this will not
* reset the previous algo every time */
static int reuse_algo = 0;
static cudnnConvolutionFwdAlgo_t prev_algo = CONV_ALGO;
#ifndef CHOOSE_ONCE
static size_t prev_img_dims[5] = {0};
static size_t prev_kern_dims[5] = {0};
reuse_algo = 1;
for (unsigned int i = 0; i < PyGpuArray_NDIM(input); i++) {
reuse_algo = (reuse_algo &&
PyGpuArray_DIM(input, i) == prev_img_dims[i]);
reuse_algo = (reuse_algo &&
PyGpuArray_DIM(kerns, i) == prev_kern_dims[i]);
}
#endif
if (!reuse_algo) {
#ifdef CHOOSE_TIME
int count;
cudnnConvolutionFwdAlgoPerf_t choice;
err = cudnnFindConvolutionForwardAlgorithm(
APPLY_SPECIFIC(_handle), APPLY_SPECIFIC(input), APPLY_SPECIFIC(kerns),
desc, APPLY_SPECIFIC(output), 1, &count, &choice);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error selecting convolution algo: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
algo = choice.algo;
#else
size_t free = 0, total = 0;
cudaError_t err2 = cudaMemGetInfo(&free, &total);
if (err2 != cudaSuccess) {
PyErr_Format(PyExc_RuntimeError, "Error when trying to find the "
"memory information on the GPU: %s\n",
cudaGetErrorString(err2));
cuda_exit(c->ctx);
return 1;
}
err = cudnnGetConvolutionForwardAlgorithm(
APPLY_SPECIFIC(_handle), APPLY_SPECIFIC(input), APPLY_SPECIFIC(kerns),
desc, APPLY_SPECIFIC(output),
CUDNN_CONVOLUTION_FWD_SPECIFY_WORKSPACE_LIMIT, free, &algo);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error selecting convolution algo: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
#endif
prev_algo = algo;
} else {
algo = prev_algo;
}
#ifdef CHOOSE_ONCE
reuse_algo = 1;
#else
for (unsigned int i = 0; i < PyGpuArray_NDIM(input); i++) {
prev_img_dims[i] = PyGpuArray_DIM(input, i);
prev_kern_dims[i] = PyGpuArray_DIM(kerns, i);
}
#endif
#endif
/* These two algos are not supported for 3d conv */
if (PyGpuArray_NDIM(input) == 5 &&
(algo == CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM ||
algo == CUDNN_CONVOLUTION_FWD_ALGO_GEMM))
algo = CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM;
#if CUDNN_VERSION > 3000
if (algo == CUDNN_CONVOLUTION_FWD_ALGO_FFT) {
int nd;
int pad[2];
int stride[2];
int upscale[2];
cudnnConvolutionMode_t mode;
err = cudnnGetConvolutionNdDescriptor(desc, 2, &nd, pad, stride,
upscale, &mode);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error getting convolution properties: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
if (stride[0] != 1 || stride[1] != 1 ||
PyGpuArray_DIM(input, 2) > 1024 || PyGpuArray_DIM(input, 3) > 1024 ||
(PyGpuArray_DIM(kerns, 2) == 1 && PyGpuArray_DIM(kerns, 3) == 1)) {
algo = CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM;
}
}
#endif
#if CUDNN_VERSION < 3000
/* cuDNN before v3 does not support kernels larger than input even
* if appropriate padding is selected. */
for (unsigned int i = 2; i < PyGpuArray_NDIM(input); i++) {
if (PyGpuArray_DIM(kerns, i) > PyGpuArray_DIM(input, i)) {
PyErr_SetString(PyExc_RuntimeError, "the current version "
"of CuDNN does not support kernels larger than the "
"inputs in any spatial dimension, even if the inputs "
"are padded such that the padded inputs are larger "
"than the kernels. Update your installation of CuDNN "
"to V3 or more recent to solve the issue.");
cuda_exit(c->ctx);
return 1;
}
}
#endif
{
size_t worksize;
gpudata *workspace;
err = cudnnGetConvolutionForwardWorkspaceSize(APPLY_SPECIFIC(_handle),
APPLY_SPECIFIC(input),
APPLY_SPECIFIC(kerns),
desc,
APPLY_SPECIFIC(output),
algo,
&worksize);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error getting worksize: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
/*
* This is less than ideal since we need to free it after (which
* introduces a synchronization point. But we don't have a module
* to place a nice get_work_mem() function in.
*/
if (worksize != 0) {
workspace = c->ops->buffer_alloc(c->ctx, worksize, NULL, 0, NULL);
if (workspace == NULL) {
PyErr_SetString(PyExc_RuntimeError,
"Could not allocate working memory");
cuda_exit(c->ctx);
return 1;
}
}
err = cudnnConvolutionForward(
APPLY_SPECIFIC(_handle),
alpha_p,
APPLY_SPECIFIC(input), PyGpuArray_DEV_DATA(input),
APPLY_SPECIFIC(kerns), PyGpuArray_DEV_DATA(kerns),
desc, algo,
worksize == 0 ? NULL : *(void **)workspace, worksize,
beta_p,
APPLY_SPECIFIC(output), PyGpuArray_DEV_DATA(*output));
if (worksize != 0)
c->ops->buffer_release(workspace);
}
cuda_exit(c->ctx);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError, "error doing operation: %s",
cudnnGetErrorString(err));
return 1;
}
return 0;
}
int APPLY_SPECIFIC(magma_eigh)(PyGpuArrayObject *A_,
PyGpuArrayObject **D,
PyGpuArrayObject **V, // may be NULL
PARAMS_TYPE *params) {
PyGpuArrayObject *A = NULL;
magma_int_t N, liwork, *iwork_data = NULL;
size_t d_dims[1], v_dims[2];
magma_uplo_t uplo;
magma_vec_t jobz;
float *w_data = NULL, *wA_data = NULL, *work_data = NULL, lwork;
int res = -1, info;
if (A_->ga.typecode != GA_FLOAT) {
PyErr_SetString(PyExc_TypeError,
"GpuMagmaEigh: Unsupported data type");
return -1;
}
// This is early to match the exit() in the fail label.
cuda_enter(params->context->ctx);
if (!GpuArray_IS_C_CONTIGUOUS(&A_->ga)) {
PyErr_SetString(PyExc_ValueError,
"GpuMagmaEigh: requires data to be C-contiguous");
goto fail;
}
if (PyGpuArray_NDIM(A_) != 2) {
PyErr_SetString(PyExc_ValueError,
"GpuMagmaEigh: matrix rank error");
goto fail;
}
if (PyGpuArray_DIM(A_, 0) != PyGpuArray_DIM(A_, 1)) {
PyErr_SetString(PyExc_ValueError,
"GpuMagmaEigh: matrix is not square");
goto fail;
}
A = pygpu_copy(A_, GA_F_ORDER);
if (A == NULL) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaEigh: failed to change to column-major order");
return -1;
}
// magma matrix eigen decomposition of a symmetric matrix
N = PyGpuArray_DIM(A, 0);
if (params->lower) {
uplo = MagmaLower;
} else {
uplo = MagmaUpper;
}
if (params->compute_v) {
jobz = MagmaVec;
} else {
jobz = MagmaNoVec;
}
if (MAGMA_SUCCESS != magma_smalloc_pinned(&w_data, N)) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaEigh: failed to allocate working memory");
goto fail;
}
if (MAGMA_SUCCESS != magma_smalloc_pinned(&wA_data, N * N)) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaEigh: failed to allocate working memory");
goto fail;
}
// query for workspace size
magma_ssyevd_gpu(jobz, uplo, N, NULL, N, NULL, NULL, N, &lwork,
-1, &liwork, -1, &info);
if (MAGMA_SUCCESS != magma_smalloc_pinned(&work_data, (size_t)lwork)) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaEigh: failed to allocate working memory");
goto fail;
}
if (MAGMA_SUCCESS != magma_imalloc_cpu(&iwork_data, liwork)) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaEigh: failed to allocate working memory");
goto fail;
}
magma_ssyevd_gpu(jobz, uplo, N, (float *)PyGpuArray_DEV_DATA(A), N, w_data,
wA_data, N, work_data, (size_t)lwork, iwork_data, liwork,
&info);
if (info > 0) {
PyErr_Format(
PyExc_RuntimeError,
"GpuMagmaEigh: %d off-diagonal elements of an didn't converge to zero",
info);
goto fail;
} else if (info < 0) {
PyErr_Format(
PyExc_RuntimeError,
"GpuMagmaEigh: magma_ssyevd_gpu argument %d has an illegal value", -info);
goto fail;
}
d_dims[0] = N;
if (theano_prep_output(D, 1, d_dims, A->ga.typecode, GA_C_ORDER, params->context) != 0){
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaEigh: failed to allocate memory for the output");
goto fail;
}
cudaMemcpy(PyGpuArray_DEV_DATA(*D), w_data, N * sizeof(float),
cudaMemcpyDeviceToDevice);
if (params->compute_v) {
*V = theano_try_copy(*V, A);
if (*V == NULL) {
PyErr_SetString(PyExc_RuntimeError,
"GpuMagmaEigh: failed to allocate memory for the output");
goto fail;
}
}
res = 0;
fail:
if (w_data != NULL)
magma_free_pinned(w_data);
if (wA_data != NULL)
magma_free_pinned(wA_data);
if (work_data != NULL)
magma_free_pinned(work_data);
if (iwork_data != NULL)
magma_free_cpu(iwork_data);
Py_XDECREF(A);
cuda_exit(params->context->ctx);
return res;
}
int APPLY_SPECIFIC(max_pool_rop)(PyGpuArrayObject *x,
PyGpuArrayObject *ex,
PyArrayObject *ws,
PyArrayObject *stride,
PyArrayObject *pad,
PyGpuArrayObject **z,
PyGpuContextObject *ctx) {
if (!GpuArray_IS_C_CONTIGUOUS(&x->ga) || !GpuArray_IS_C_CONTIGUOUS(&ex->ga))
{
PyErr_Format(PyExc_ValueError,
"GpuMaxPoolRop: requires data to be C-contiguous");
return 1;
}
size_t ndims = PyArray_DIM(ws, 0);
if (PyGpuArray_NDIM(x) != ndims + 2 || PyGpuArray_NDIM(ex) != ndims + 2)
{
PyErr_SetString(PyExc_ValueError, "GpuMaxPoolRop: rank error");
return 1;
}
// prepare output
const size_t* x_dims = PyGpuArray_DIMS(x);
size_t z_dims[5]; // avoid warning if use 2 + nd
size_t w[3];
size_t s[3];
size_t p[3]; z_dims[0] = x_dims[0]; z_dims[1] = x_dims[1];
int nonzero_padding = 0;
for (int i = 0; i < ndims; i++) {
w[i] = *((npy_int64*)PyArray_GETPTR1(ws, i));
s[i] = *((npy_int64*)PyArray_GETPTR1(stride, i));
p[i] = *((npy_int64*)PyArray_GETPTR1(pad, i));
z_dims[2 + i] = OUTPUT_DIMS(x_dims[2 + i] + 2*p[i], w[i], s[i]);
if (p[i] > 0) {
nonzero_padding = 1;
}
}
if (!IGNORE_BORDER && nonzero_padding) {
PyErr_SetString(PyExc_ValueError,
"GpuMaxPoolRop: padding works only with ignore_border=True");
return 1;
}
if (theano_prep_output(z, PyGpuArray_NDIM(ex), z_dims,
ex->ga.typecode, GA_C_ORDER, ctx) != 0)
{
PyErr_SetString(PyExc_RuntimeError,
"GpuMaxPoolRop: failed to allocate memory");
return 1;
}
{
// scope for running kernel
int err;
if (ndims == 2) {
size_t num_kernels = z_dims[0] * z_dims[1] * z_dims[2] * z_dims[3];
err = max_pool2d_rop_kernel_scall(1, &num_kernels, 0, num_kernels,
z_dims[0], z_dims[1], z_dims[2], z_dims[3],
x_dims[2], x_dims[3],
x->ga.data, ex->ga.data,
w[0], w[1], s[0], s[1], p[0], p[1],
(*z)->ga.data);
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuMaxPoolRop: max_pool2d_rop_kernel %s.",
GpuKernel_error(&k_max_pool2d_rop_kernel, err));
return 1;
}
}
else if (ndims == 3) {
size_t num_kernels = z_dims[0] * z_dims[1] * z_dims[2] * z_dims[3] * z_dims[4];
err = max_pool3d_rop_kernel_scall(1, &num_kernels, 0, num_kernels,
z_dims[0], z_dims[1], z_dims[2], z_dims[3], z_dims[4],
x_dims[2], x_dims[3], x_dims[4],
x->ga.data, ex->ga.data,
w[0], w[1], w[2], s[0], s[1], s[2],
p[0], p[1], p[2], (*z)->ga.data);
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuMaxPoolRop: max_pool3d_rop_kernel %s.",
GpuKernel_error(&k_max_pool2d_rop_kernel, err));
return 1;
}
}
}
return 0;
}
// Theano op code
// Authors: Arjun Jain, Frederic Bastien, Jan Schluter
// Reference code: https://github.com/BVLC/caffe/blob/master/src/caffe/layers/conv_layer.cu
// and https://github.com/torch/cunn/blob/master/SpatialConvolutionMM.cu
// Adaptation for 3d
PyGpuArrayObject* corr3dMM(PyGpuArrayObject *const bottom,
PyGpuArrayObject *const weight,
PyGpuArrayObject *const top,
const size_t direction,
const size_t dH = 1,
const size_t dW = 1,
const size_t dD = 1,
const size_t dilH = 1,
const size_t dilW = 1,
const size_t dilD = 1,
const size_t padH = 0,
const size_t padW = 0,
const size_t padD = 0)
{
if (PyGpuArray_NDIM(bottom) != 5)
{
PyErr_SetString(PyExc_ValueError, "GpuCorr3dMM requires bottom of 5D");
return NULL;
}
if (!GpuArray_IS_C_CONTIGUOUS(&bottom->ga))
{
PyErr_Format(PyExc_ValueError,
"GpuCorr3dMM requires bottom to be C-contiguous, "
"but strides are: %ld %ld %ld %ld %ld\n",
PyGpuArray_STRIDES(bottom)[0],
PyGpuArray_STRIDES(bottom)[1],
PyGpuArray_STRIDES(bottom)[2],
PyGpuArray_STRIDES(bottom)[3],
PyGpuArray_STRIDES(bottom)[4]);
return NULL;
}
if (PyGpuArray_NDIM(weight) != 5)
{
PyErr_SetString(PyExc_ValueError, "GpuCorr3dMM requires weight of 5D");
return NULL;
}
if (!GpuArray_IS_C_CONTIGUOUS(&weight->ga))
{
PyErr_Format(PyExc_ValueError,
"GpuCorr3dMM requires weight to be C-contiguous, "
"but strides are: %ld %ld %ld %ld %ld\n",
PyGpuArray_STRIDES(weight)[0],
PyGpuArray_STRIDES(weight)[1],
PyGpuArray_STRIDES(weight)[2],
PyGpuArray_STRIDES(weight)[3],
PyGpuArray_STRIDES(weight)[4]);
return NULL;
}
if (PyGpuArray_NDIM(top) != 5)
{
PyErr_SetString(PyExc_ValueError, "GpuCorr3dMM requires top of 5D");
return NULL;
}
if (!GpuArray_IS_C_CONTIGUOUS(&top->ga))
{
PyErr_Format(PyExc_ValueError,
"GpuCorr3dMM requires top to be C-contiguous, "
"but strides are: %ld %ld %ld %ld %ld\n",
PyGpuArray_STRIDES(top)[0],
PyGpuArray_STRIDES(top)[1],
PyGpuArray_STRIDES(top)[2],
PyGpuArray_STRIDES(top)[3],
PyGpuArray_STRIDES(top)[4]);
return NULL;
}
// Extract some shape information for later and check shape consistency
// bottom: (batchSize, nChannels, bottomHeight, bottomWidth, bottomDepth)
const size_t batchSize = PyGpuArray_DIMS(bottom)[0];
const size_t nChannels = PyGpuArray_DIMS(bottom)[1];
const size_t bottomHeight = PyGpuArray_DIMS(bottom)[2];
const size_t bottomWidth = PyGpuArray_DIMS(bottom)[3];
const size_t bottomDepth = PyGpuArray_DIMS(bottom)[4];
// weights: (nFilters, nChannels, rows, columns, slices)
const size_t nFilters = PyGpuArray_DIMS(weight)[0];
const size_t kH = PyGpuArray_DIMS(weight)[2];
const size_t kW = PyGpuArray_DIMS(weight)[3];
const size_t kD = PyGpuArray_DIMS(weight)[4];
if (nChannels != PyGpuArray_DIMS(weight)[1]) {
PyErr_SetString(PyExc_ValueError,
"GpuCorr3dMM images and kernel must have the same stack size\n");
return NULL;
}
// implicit dilated filter
const size_t dil_kH = (kH - 1) * dilH + 1;
const size_t dil_kW = (kW - 1) * dilW + 1;
const size_t dil_kD = (kD - 1) * dilD + 1;
// top: (batchSize, nFilters, topHeight, topWidth, topDepth)
const size_t topHeightNoDH = (bottomHeight + 2*padH - dil_kH);
const size_t topWidthNoDW = (bottomWidth + 2*padW - dil_kW);
const size_t topDepthNoDD = (bottomDepth + 2*padD - dil_kD);
// the above values might be negative so we need to use Python-like
// flooring integer division to be compatible with get_conv_output.
// note: this macro implements Python's // for negative x only
#define _CONV_FLOORDIV_X(x,y) ((x < 0) ? (- ((-x) / y) - (((-x) % y) == 0 ? 0 : 1)) : (x / y))
const size_t topHeight = _CONV_FLOORDIV_X(topHeightNoDH, dH) + 1;
const size_t topWidth = _CONV_FLOORDIV_X(topWidthNoDW, dW) + 1;
const size_t topDepth = _CONV_FLOORDIV_X(topDepthNoDD, dD) + 1;
#undef _CONV_FLOORDIV
if (batchSize != PyGpuArray_DIMS(top)[0] ||
nFilters != PyGpuArray_DIMS(top)[1] ||
topHeight != PyGpuArray_DIMS(top)[2] ||
topWidth != PyGpuArray_DIMS(top)[3] ||
topDepth != PyGpuArray_DIMS(top)[4]) {
PyErr_Format(PyExc_ValueError,
"GpuCorr3dMM shape inconsistency:\n"
" bottom shape: %ld %ld %ld %ld %ld\n"
" weight shape: %ld %ld %ld %ld %ld\n"
" top shape: %ld %ld %ld %ld %ld (expected %ld %ld %ld %ld %ld)\n",
batchSize, nChannels, bottomHeight, bottomWidth, bottomDepth,
nFilters, nChannels, kH, kW, kD,
PyGpuArray_DIMS(top)[0], PyGpuArray_DIMS(top)[1],
PyGpuArray_DIMS(top)[2], PyGpuArray_DIMS(top)[3], PyGpuArray_DIMS(top)[4],
batchSize, nFilters, topHeight, topWidth, topDepth);
return NULL;
}
int err = gpublas_setup(bottom->context->ctx);
if (err != GA_NO_ERROR) {
PyErr_SetString(PyExc_RuntimeError, "Can't setup blas");
return NULL;
}
// Create temporary columns
size_t col_dim[2];
col_dim[0] = nChannels * kW * kH * kD;
col_dim[1] = topHeight * topWidth * topDepth;
PyGpuArrayObject* col = (PyGpuArrayObject*)pygpu_empty(2, col_dim,
bottom->ga.typecode,
GA_C_ORDER,
bottom->context,
Py_None);
if (NULL == col)
{
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM failed to allocate working memory of %ld x %ld\n",
col_dim[0], col_dim[1]);
return NULL;
}
// Define some useful variables
const size_t bottom_stride = PyGpuArray_STRIDES(bottom)[0] / gpuarray_get_elsize(bottom->ga.typecode);
const size_t top_stride = PyGpuArray_STRIDES(top)[0] / gpuarray_get_elsize(top->ga.typecode);
const size_t K_ = col_dim[0];
const size_t N_ = col_dim[1];
const size_t M_ = nFilters;
PyGpuArrayObject *output;
if (direction == 0) { // forward pass
output = top;
if (batchSize == 0 || nChannels == 0 || nFilters == 0) {
err = GpuArray_memset(&output->ga, 0);
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM could not fill the output with zeros: %d", err);
Py_DECREF(col);
return NULL;
}
Py_DECREF(col);
return output;
}
// valid correlation: im3d2col, then gemm
// Iterate over batch
for (size_t n = 0; n < batchSize; n++) {
// First, im3d2col
err = im3d2col(
bottom->ga.data, n * bottom_stride, nChannels, bottomHeight,
bottomWidth, bottomDepth, kH, kW, kD, dilH, dilW, dilD,
padH, padW, padD, dH, dW, dD, col->ga.data);
if (err != GA_NO_ERROR) {
Py_DECREF(col);
return NULL;
}
// Second, gemm
switch (col->ga.typecode) {
case GA_FLOAT:
err = gpublas_sgemm(cb_fortran, cb_no_trans, cb_no_trans,
N_, M_, K_, 1,
col->ga.data, 0, N_,
weight->ga.data, 0, K_,
0,
top->ga.data, n * top_stride, N_);
break;
case GA_DOUBLE:
err = gpublas_dgemm(cb_fortran, cb_no_trans, cb_no_trans,
N_, M_, K_, 1,
col->ga.data, 0, N_,
weight->ga.data, 0, K_,
0,
top->ga.data, n * top_stride, N_);
break;
case GA_HALF:
err = gpublas_hgemm(cb_fortran, cb_no_trans, cb_no_trans,
N_, M_, K_, 1,
col->ga.data, 0, N_,
weight->ga.data, 0, K_,
0,
top->ga.data, n * top_stride, N_);
break;
default:
err = GA_UNSUPPORTED_ERROR;
}
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM forward encountered an error running gemm.");
Py_DECREF(col);
return NULL;
}
}
}
else if (direction == 1) { // backprop wrt. weights
output = weight;
if (batchSize == 0 || nChannels == 0 || nFilters == 0) {
err = GpuArray_memset(&output->ga, 0);
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM grad wrt. weights could not fill the output with zeros: %d", err);
Py_DECREF(col);
return NULL;
}
Py_DECREF(col);
return output;
}
// valid convolution: im3col, then gemm
// Iterate over batch
for (size_t n = 0; n < batchSize; n++) {
// First, im3d2col
err = im3d2col(
bottom->ga.data, n * bottom_stride, nChannels, bottomHeight,
bottomWidth, bottomDepth, kH, kW, kD, dilH, dilW, dilD,
padH, padW, padD, dH, dW, dD, col->ga.data);
if (err != GA_NO_ERROR) {
Py_DECREF(col);
return NULL;
}
// Second, gemm
// Note that we accumulate into weight. We do so by setting beta = 0
// for the first iteration and beta = 1 for subsequent ones. (This
// is faster than setting weight to all zeros before the loop.)
switch (col->ga.typecode) {
case GA_FLOAT:
err = gpublas_sgemm(cb_fortran, cb_trans, cb_no_trans,
K_, M_, N_, 1,
col->ga.data, 0, N_,
top->ga.data, n * top_stride, N_,
(n == 0) ? 0 : 1,
weight->ga.data, 0, K_);
break;
case GA_DOUBLE:
err = gpublas_dgemm(cb_fortran, cb_trans, cb_no_trans,
K_, M_, N_, 1,
col->ga.data, 0, N_,
top->ga.data, n * top_stride, N_,
(n == 0) ? 0 : 1,
weight->ga.data, 0, K_);
break;
case GA_HALF:
err = gpublas_hgemm(cb_fortran, cb_trans, cb_no_trans,
K_, M_, N_, 1,
col->ga.data, 0, N_,
top->ga.data, n * top_stride, N_,
(n == 0) ? 0 : 1,
weight->ga.data, 0, K_);
break;
default:
err = GA_UNSUPPORTED_ERROR;
}
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM grad weights encountered an error running gemm.");
Py_DECREF(col);
return NULL;
}
}
if (batchSize == 0) {
err = GpuArray_memset(&weight->ga, 0);
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM grad weights could not fill the output with zeros: %d", err);
Py_DECREF(col);
return NULL;
}
}
}
else if (direction == 2) { // backprop wrt. inputs
output = bottom;
if (batchSize == 0 || nChannels == 0 || nFilters == 0) {
err = GpuArray_memset(&output->ga, 0);
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM grad wrt. inputs could not fill the output with zeros: %d", err);
Py_DECREF(col);
return NULL;
}
Py_DECREF(col);
return output;
}
// full convolution: gemm, then col2im3d
// Iterate over batch
for (size_t n = 0; n < batchSize; n++) {
// gemm into columns
switch (top->ga.typecode) {
case GA_FLOAT:
err = gpublas_sgemm(cb_fortran, cb_no_trans, cb_trans,
N_, K_, M_, 1,
top->ga.data, n * top_stride, N_,
weight->ga.data, 0, K_,
0,
col->ga.data, 0, N_);
break;
case GA_DOUBLE:
err = gpublas_dgemm(cb_fortran, cb_no_trans, cb_trans,
N_, K_, M_, 1,
top->ga.data, n * top_stride, N_,
weight->ga.data, 0, K_,
0,
col->ga.data, 0, N_);
break;
case GA_HALF:
err = gpublas_hgemm(cb_fortran, cb_no_trans, cb_trans,
N_, K_, M_, 1,
top->ga.data, n * top_stride, N_,
weight->ga.data, 0, K_,
0,
col->ga.data, 0, N_);
break;
default:
err = GA_UNSUPPORTED_ERROR;
}
if (err != GA_NO_ERROR) {
PyErr_Format(PyExc_RuntimeError,
"GpuCorr3dMM grad inputs encountered an error running gemm.");
Py_DECREF(col);
return NULL;
}
// col2im3d back to the data
err = col2im3d(col->ga.data, nChannels,
bottomHeight, bottomWidth, bottomDepth,
kH, kW, kD, dilH, dilW, dilD, padH, padW, padD,
dH, dW, dD, bottom->ga.data, n * bottom_stride);
if (err != GA_NO_ERROR) {
Py_DECREF(col);
return NULL;
}
}
}
// Free temporary columns
Py_DECREF(col);
// Note that we don't change the refcount of the output matrix here. Output
// (re)allocation and refcounting is done in BaseGpuCorr3dMM.c_code_helper();
// in here output is just aliased to one of bottom, weights, or top.
return output;
}
int
APPLY_SPECIFIC(conv_gi)(PyGpuArrayObject *kerns, PyGpuArrayObject *output,
PyGpuArrayObject *im,
cudnnConvolutionDescriptor_t desc,
double alpha, double beta, PyGpuArrayObject **input,
PyGpuContextObject *c) {
cudnnStatus_t err = CUDNN_STATUS_SUCCESS;
float af = alpha, bf = beta;
void *alpha_p;
void *beta_p;
if (PyGpuArray_DIMS(im)[1] != PyGpuArray_DIMS(kerns)[1]) {
PyErr_SetString(PyExc_ValueError, "images and kernel must have the same "
"stack size");
return 1;
}
if (c_set_tensorNd(output, APPLY_SPECIFIC(output)) == -1)
return 1;
if (c_set_filter(kerns, APPLY_SPECIFIC(kerns)) == -1)
return 1;
switch (im->ga.typecode) {
case GA_DOUBLE:
alpha_p = (void *)α
beta_p = (void *)β
break;
case GA_FLOAT:
case GA_HALF:
alpha_p = (void *)⁡
beta_p = (void *)&bf;
break;
default:
PyErr_SetString(PyExc_TypeError, "Unsupported type in convolution");
return 1;
}
#ifdef CONV_INPLACE
Py_XDECREF(*input);
*input = im;
Py_INCREF(*input);
#else
if (theano_prep_output(input, PyGpuArray_NDIM(im), PyGpuArray_DIMS(im),
im->ga.typecode, GA_C_ORDER, c) != 0)
return 1;
if (beta != 0.0 && pygpu_move(*input, im))
return 1;
#endif
if (c_set_tensorNd(*input, APPLY_SPECIFIC(input)) == -1)
return 1;
cudnnConvolutionBwdDataAlgo_t algo = CONV_ALGO;
cuda_enter(c->ctx);
#ifdef CHOOSE_ALGO
static int reuse_algo = 0;
static cudnnConvolutionBwdDataAlgo_t prev_algo = CONV_ALGO;
#ifndef CHOOSE_ONCE
static size_t prev_kern_dims[5] = {0};
static size_t prev_top_dims[5] = {0};
reuse_algo = 1;
for (unsigned int i = 0; i < PyGpuArray_NDIM(kerns); i++) {
reuse_algo = (reuse_algo &&
PyGpuArray_DIM(kerns, i) == prev_kern_dims[i]);
reuse_algo = (reuse_algo &&
PyGpuArray_DIM(output, i) == prev_top_dims[i]);
}
#endif
if (!reuse_algo) {
#ifdef CHOOSE_TIME
int count;
cudnnConvolutionBwdDataAlgoPerf_t choice;
err = cudnnFindConvolutionBackwardDataAlgorithm(
APPLY_SPECIFIC(_handle), APPLY_SPECIFIC(input), APPLY_SPECIFIC(output), desc,
APPLY_SPECIFIC(kerns), 1, &count, &choice);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError, "error selecting convolution algo: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
algo = choice.algo;
#else
size_t free = 0, total = 0;
cudaError_t err2 = cudaMemGetInfo(&free, &total);
if (err2 != cudaSuccess) {
cudaGetLastError();
PyErr_Format(PyExc_RuntimeError, "Error when trying to find the memory "
"information on the GPU: %s\n", cudaGetErrorString(err2));
cuda_exit(c->ctx);
return 1;
}
err = cudnnGetConvolutionBackwardDataAlgorithm(
APPLY_SPECIFIC(_handle), APPLY_SPECIFIC(input), APPLY_SPECIFIC(output),
desc, APPLY_SPECIFIC(kerns),
CUDNN_CONVOLUTION_BWD_DATA_SPECIFY_WORKSPACE_LIMIT, free, &algo);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError, "error selecting convolution algo: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
#endif
prev_algo = algo;
} else {
algo = prev_algo;
}
#ifdef CHOOSE_ONCE
reuse_algo = 1;
#else
for (unsigned int i = 0; i < PyGpuArray_NDIM(kerns); i++) {
prev_kern_dims[i] = PyGpuArray_DIM(kerns, i);
prev_top_dims[i] = PyGpuArray_DIM(output, i);
}
#endif
#endif
#if CUDNN_VERSION > 3000
if (algo == CUDNN_CONVOLUTION_BWD_DATA_ALGO_FFT) {
int nd;
int pad[2];
int stride[2];
int upscale[2];
cudnnConvolutionMode_t mode;
err = cudnnGetConvolutionNdDescriptor(desc, 2, &nd, pad, stride,
upscale, &mode);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError,
"error getting convolution properties: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
if (stride[0] != 1 || stride[1] != 1 ||
PyGpuArray_DIM(*input, 2) > 1024 || PyGpuArray_DIM(*input, 3) > 1024 ||
(PyGpuArray_DIM(kerns, 2) == 1 && PyGpuArray_DIM(kerns, 3) == 1)) {
algo = CUDNN_CONVOLUTION_BWD_DATA_ALGO_0;
}
}
#endif
size_t worksize;
gpudata *workspace;
err = cudnnGetConvolutionBackwardDataWorkspaceSize(
APPLY_SPECIFIC(_handle), APPLY_SPECIFIC(kerns), APPLY_SPECIFIC(output), desc,
APPLY_SPECIFIC(input), algo, &worksize);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError, "error getting worksize: %s",
cudnnGetErrorString(err));
cuda_exit(c->ctx);
return 1;
}
if (worksize != 0) {
workspace = c->ops->buffer_alloc(c->ctx, worksize, NULL, 0, NULL);
if (workspace == NULL) {
PyErr_SetString(PyExc_RuntimeError,
"Could not allocate working memory");
cuda_exit(c->ctx);
return 1;
}
}
cuda_wait(kerns->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_wait(output->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_wait((*input)->ga.data, GPUARRAY_CUDA_WAIT_WRITE);
err = cudnnConvolutionBackwardData_v3(
APPLY_SPECIFIC(_handle),
alpha_p,
APPLY_SPECIFIC(kerns), PyGpuArray_DEV_DATA(kerns),
APPLY_SPECIFIC(output), PyGpuArray_DEV_DATA(output),
desc, algo, worksize == 0 ? NULL : *(void **)workspace, worksize,
beta_p,
APPLY_SPECIFIC(input), PyGpuArray_DEV_DATA(*input));
if (worksize != 0)
c->ops->buffer_release(workspace);
cuda_record(kerns->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_record(output->ga.data, GPUARRAY_CUDA_WAIT_READ);
cuda_record((*input)->ga.data, GPUARRAY_CUDA_WAIT_WRITE);
cuda_exit(c->ctx);
if (err != CUDNN_STATUS_SUCCESS) {
PyErr_Format(PyExc_RuntimeError, "error doing operation: %s",
cudnnGetErrorString(err));
return 1;
}
return 0;
}