提交 a8863a8d 编写于 作者: D dangqingqing

Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cross_entropy

上级 8f6c8780 0d9846f3
develop 1.8.5 2.0.1-rocm-post 2.4.1 Ligoml-patch-1 OliverLPH-patch-1 OliverLPH-patch-2 PaddlePM-patch-1 PaddlePM-patch-2 ZHUI-patch-1 add_default_att add_kylinv10 add_model_benchmark_ci add_some_yaml_config addfile all_new_design_exec ascendrc ascendrelease bugfix-eval-frame-leakgae cherry-pick-fix-customOP-random-fail cherry_undefined_var compile_windows cp_2.4_fix_numpy delete_2.0.1-rocm-post delete_add_default_att delete_all_new_design_exec delete_ascendrc delete_compile_windows delete_delete_addfile delete_disable_iterable_dataset_unittest delete_fix_dataloader_memory_leak delete_fix_imperative_dygraph_error delete_fix_retry_ci delete_fix_undefined_var delete_improve_sccache delete_incubate/lite delete_paddle_tiny_install delete_paralleltest delete_prv-disable-more-cache delete_revert-31068-fix_conv3d_windows delete_revert-31562-mean delete_revert-33630-bug-fix delete_revert-34159-add_npu_bce_logical_dev delete_revert-34910-spinlocks_for_allocator delete_revert-35069-revert-34910-spinlocks_for_allocator delete_revert-36057-dev/read_flags_in_ut dingjiaweiww-patch-1 disable_iterable_dataset_unittest dy2static enable_eager_model_test final_state_gen_python_c final_state_intermediate fix-numpy-issue fix-run-program-grad-node-mem fix_check fix_concat_slice fix_custom_device_copy_sync fix_dataloader_memory_leak fix_dlpack_for fix_imperative_dygraph_error fix_newexe_gc fix_npu_ci fix_op_flops fix_retry_ci fix_rnn_docs fix_tensor_type fix_undefined_var fix_var_stop_gradient_error fixiscan fixiscan1 fixiscan2 fixiscan3 github/fork/123malin/netifaces github/fork/123malin/tdm_abacus github/fork/AshburnLee/dev_unique github/fork/ForFishes/fix_memory_matmul github/fork/ForFishes/rm_fluid github/fork/LielinJiang/move-2.0-api github/fork/LielinJiang/visual-dl-cb github/fork/LiuChiachi/add-transformer-generate-square-subsequent-mask-api github/fork/LiuChiachi/fix-example-code-for-hapi-Model github/fork/LiuChiachi/remove-input-requirment-in-dygraph-Model github/fork/MrChengmo/fix_ps_profiler github/fork/MrChengmo/update_ps_heter github/fork/PWhiddy/patch-1 github/fork/Shixiaowei02/dev/save_load_upgrade github/fork/TCChenlong/fix_hapi github/fork/TCChenlong/fix_inden github/fork/Thunderbrook/xpu_slice github/fork/XieYunshen/disable_ut_test_parallel_executor_fetch_isolated_var github/fork/XieYunshen/disable_ut_test_parallel_executor_fetch_isolated_var_2 github/fork/XieYunshen/disable_ut_test_parallel_executor_fetch_isolated_var_3 github/fork/XieYunshen/timeout_20S_ut github/fork/ZeyuChen/remove-nltk github/fork/arlesniak/arlesniak/selective__mkldnn_flags github/fork/baiyfbupt/code_doc_mig github/fork/chalsliu/set_timeout github/fork/chen-zhiyu/develop github/fork/chenwhql/ci/try_to_find_test_buffer_shared_memory_reuse_pass_error github/fork/chenwhql/dygraph/remove_scale_loss_and_apply_collective_grads github/fork/chenwhql/saveload/add_get_inference_program github/fork/chenwhql/saveload/remove_save_load_config github/fork/cryoco/pass-compatibility-trt github/fork/danleifeng/isempty_api2.0 github/fork/frankwhzhang/api_transfer github/fork/hbwx24/error_msg/cuda_kernel_error_msg github/fork/heavengate/cherry_yolo_box github/fork/heavengate/update_yolo_box github/fork/iclementine/rnn_fix github/fork/iducn/testestse github/fork/jczaja/prv-25537-fix github/fork/jeff41404/release/1.8 github/fork/jiweibo/api_2.0 github/fork/jiweibo/fix_lite_resnet50_test github/fork/juncaipeng/fix_doc_1 github/fork/lfchener/sample_code github/fork/littletomatodonkey/fix_reg_doc github/fork/liym27/dy2stat_update_assign_to_rc20 github/fork/luotao1/profiler_ut github/fork/mapingshuo/add_wait github/fork/mapingshuo/doc_2.0 github/fork/mapingshuo/zero-0.5 github/fork/miraiwk/dev github/fork/pangyoki/add-Categorical-class-branch github/fork/pangyoki/add-multinomial-op-branch github/fork/pangyoki/fix-test_distritbution-CI github/fork/qjing666/doublegrad github/fork/qjing666/fix_hdfs_download github/fork/sandyhouse/add_gather_etc github/fork/sandyhouse/add_send_recv_alltoall_etc github/fork/sandyhouse/pipeline_exe_run github/fork/seiriosPlus/feature/large_scale_kv_save_delta github/fork/seiriosPlus/fix/paddle_errors_fix github/fork/seiriosPlus/fix/paddle_op_errors github/fork/shangzhizhou/fix_test_activation_op_random_bug github/fork/smallv0221/yxp0924 github/fork/smallv0221/yxp0925 github/fork/swtkiwi/del-matplotlib github/fork/tianshuo78520a/kunlun_test github/fork/tianshuo78520a/update_dockerfile github/fork/wanghaoshuang/bert_fuse github/fork/wanghaoshuang/label_smooth github/fork/wanghuancoder/develop_CUDASynchronize github/fork/wanghuancoder/develop_Layer_doc github/fork/wanghuancoder/develop_ParameterList_doc github/fork/wanghuancoder/develop_Sequential_doc github/fork/wanghuancoder/develop_bilinear_tensor_product github/fork/wanghuancoder/develop_coverage_build_sh github/fork/wanghuancoder/develop_in_dynamic_mode_doc github/fork/wanghuancoder/develop_unique_name_doc github/fork/wangxicoding/fleet_meta_combine github/fork/wawltor/error_message_fix_5 github/fork/willthefrog/remove_l2_norm github/fork/windstamp/momentum_op github/fork/windstamp/mv_op_5 github/fork/windstamp/normal_api github/fork/wojtuss/wojtuss/fusion_gru_quantization github/fork/wojtuss/wojtuss/quantization-with-shift github/fork/wzzju/fix_err_info github/fork/wzzju/pure_fp16 github/fork/xiemoyuan/op_error_message github/fork/xiemoyuan/optimize_error_message github/fork/yaoxuefeng6/fix_doc github/fork/yaoxuefeng6/mod_dataset_v2 github/fork/yongqiangma/lod github/fork/ysh329/fix-clip-by-norm-error github/fork/ysh329/fix-error-clip-by-value github/fork/yukavio/error_info github/fork/zhangting2020/conv_filter_grad github/fork/zhangting2020/is_compile_with_cuda github/fork/zhangting2020/place_doc github/fork/zhangting2020/program github/fork/zhhsplendid/fix_any github/fork/zhhsplendid/refine_api2 github/fork/zhhsplendid/refine_api2_test github/fork/zhhsplendid/refine_api_test_ptb_lm github/fork/zhhsplendid/refine_api_test_resnet github/fork/zhhsplendid/refine_api_test_simnet github/fork/zhiqiu/dev/refine_initializer github/fork/zhiqiu/dev/remove_inplace_argument github/fork/zlsh80826/nvinfer_plugin_var_len_cuda11 hack_event improve_sccache incuabte/new_frl incubate/frl_train_eval incubate/infrt incubate/lite incubate/new_frl incubate/new_frl_rc incubate/stride inplace_addto layer_norm make_flag_adding_easier master matmul_double_grad move_embedding_to_phi move_histogram_to_pten move_sgd_to_phi move_slice_to_pten move_temporal_shift_to_phi move_yolo_box_to_phi npu_fix_alloc numel operator_opt paddle_tiny_install paralleltest pass-compile-eval-frame preln_ernie prv-disable-more-cache prv-md-even-more prv-onednn-2.5 prv-reshape-mkldnn-ut2 pten_tensor_refactor release-deleted/2.5 release-rc/2.5 release/0.11.0 release/0.12.0 release/0.13.0 release/0.14.0 release/0.15.0 release/1.0.0 release/1.1 release/1.2 release/1.3 release/1.4 release/1.5 release/1.6 release/1.7 release/1.8 release/2.0 release/2.0-alpha release/2.0-beta release/2.0-rc release/2.0-rc1 release/2.1 release/2.2 release/2.3 release/2.3-fc-ernie-fix release/2.4 release/2.5 release/lite-0.1 release/llm_2.5 revert-24981-add_device_attr_for_regulization revert-26856-strategy_example2 revert-27520-disable_pr revert-31068-fix_conv3d_windows revert-31562-mean revert-32290-develop-hardlabel revert-33037-forci revert-33475-fix_cifar_label_dimension revert-33630-bug-fix revert-34159-add_npu_bce_logical_dev revert-34406-add_copy_from_tensor revert-34910-spinlocks_for_allocator revert-35069-revert-34910-spinlocks_for_allocator revert-36057-dev/read_flags_in_ut revert-36201-refine_fast_threaded_ssa_graph_executor revert-36985-add_license revert-37318-refactor_dygraph_to_eager revert-37926-eager_coreops_500 revert-37956-revert-37727-pylayer_support_tuple revert-38100-mingdong revert-38301-allocation_rearrange_pr revert-38703-numpy_bf16_package_reupload revert-38732-remove_useless_header_in_elementwise_mul_grad revert-38959-Reduce_Grad revert-39143-adjust_empty revert-39227-move_trace_op_to_pten revert-39268-dev/remove_concat_fluid_kernel revert-40170-support_partial_grad revert-41056-revert-40727-move_some_activaion_to_phi revert-41065-revert-40993-mv_ele_floordiv_pow revert-41068-revert-40790-phi_new revert-41944-smaller_inference_api_test revert-42149-do-not-reset-default-stream-for-stream-safe-cuda-allocator revert-43155-fix_ut_tempfile revert-43882-revert-41944-smaller_inference_api_test revert-45808-phi/simplify_size_op revert-46827-deform_comment revert-47325-remove_cudnn_hardcode revert-47645-add_npu_storage_dims revert-48815-set_free_when_no_cache_hit_default_value_true revert-49499-test_ninja_on_ci revert-49654-prim_api_gen revert-49673-modify_get_single_cov revert-49763-fix_static_composite_gen revert-50158-fix_found_inf_bug_for_custom_optimizer revert-50188-refine_optimizer_create_accumulators revert-50335-fix_optminizer_set_auxiliary_var_bug revert-51676-flag_delete revert-51850-fix_softmaxce_dev revert-52175-dev_peak_memory revert-52186-deve revert-52523-test_py38 revert-52912-develop revert-53248-set_cmake_policy revert-54029-fix_windows_compile_bug revert-54068-support_translating_op_attribute revert-54214-modify_cmake_dependencies revert-54370-offline_pslib revert-54391-fix_cmake_md5error revert-54411-fix_cpp17_compile revert-54466-offline_pslib revert-54480-cmake-rocksdb revert-55568-fix_BF16_bug1 revert-56328-new_ir_support_vector_type_place_transfer revert-56366-fix_openssl_bug revert-56545-revert-56366-fix_openssl_bug revert-56620-fix_new_ir_ocr_bug revert-56925-check_inputs_grad_semantic revert-57005-refine_stride_flag rocm_dev_0217 sd_conv_linear_autocast semi-auto/rule-base support-0D-sort support_weight_transpose test_benchmark_ci test_feature_precision_test_c test_for_Filtetfiles test_model_benchmark test_model_benchmark_ci zhiqiu-patch-1 v2.5.1 v2.5.0 v2.5.0-rc1 v2.5.0-rc0 v2.4.2 v2.4.1 v2.4.0 v2.4.0-rc0 v2.3.2 v2.3.1 v2.3.0 v2.3.0-rc0 v2.2.2 v2.2.1 v2.2.0 v2.2.0-rc0 v2.2.0-bak0 v2.1.3 v2.1.2 v2.1.1 v2.1.0 v2.1.0-rc0 v2.0.2 v2.0.1 v2.0.0 v2.0.0-rc1 v2.0.0-rc0 v2.0.0-beta0 v2.0.0-alpha0 v1.8.5 v1.8.4 v1.8.3 v1.8.2 v1.8.1 v1.8.0 v1.7.2 v1.7.1 v1.7.0 v1.6.3 v1.6.2 v1.6.1 v1.6.0 v1.6.0-rc0 v1.5.2 v1.5.1 v1.5.0 v1.4.1 v1.4.0 v1.3.2 v1.3.1 v1.3.0 v1.2.1 v1.2.0 v1.1.0 v1.0.2 v1.0.1 v1.0.0 v1.0.0-rc0 v0.15.0 v0.15.0-rc0 v0.14.0 v0.13.0 v0.12.0 v0.11.1a2 v0.11.1a1 v0.11.0 lite-v0.1
7 合并请求!11636[IMPORTANT] MKLDNN layout: Support for sum operator,!8482Release/0.11.0,!8190Release/0.11.0,!8189Release/0.11.0,!6633给线性回归的get-started代码加上了预测的示例~~,!4615Feature/tensor array add python binding,!3501Implement GPU kernel for cross entropy operator.
......@@ -2,7 +2,7 @@ if(NOT WITH_GPU)
return()
endif()
set(CUDNN_ROOT "" CACHE PATH "CUDNN ROOT")
set(CUDNN_ROOT "/usr" CACHE PATH "CUDNN ROOT")
find_path(CUDNN_INCLUDE_DIR cudnn.h
PATHS ${CUDNN_ROOT} ${CUDNN_ROOT}/include
$ENV{CUDNN_ROOT} $ENV{CUDNN_ROOT}/include ${CUDA_TOOLKIT_INCLUDE}
......
......@@ -19,8 +19,13 @@ limitations under the License. */
#include <memory> // for unique_ptr
#include <mutex> // for call_once
#include "glog/logging.h"
#include "paddle/memory/detail/buddy_allocator.h"
#include "paddle/memory/detail/system_allocator.h"
#include "paddle/platform/gpu_info.h"
DECLARE_double(fraction_of_gpu_memory_to_use);
namespace paddle {
namespace memory {
......@@ -80,6 +85,11 @@ BuddyAllocator* GetGPUBuddyAllocator(int gpu_id) {
platform::GpuMinChunkSize(),
platform::GpuMaxChunkSize()));
}
VLOG(3) << "\n\nNOTE: each GPU device use "
<< FLAGS_fraction_of_gpu_memory_to_use * 100 << "% of GPU memory.\n"
<< "You can set environment variable '"
<< platform::kEnvFractionGpuMemoryToUse
<< "' to change the fraction of GPU usage.\n\n";
});
platform::SetDeviceId(gpu_id);
......
......@@ -14,7 +14,6 @@ limitations under the License. */
#pragma once
#include "paddle/platform/gpu_info.h"
#include "paddle/platform/place.h"
namespace paddle {
......
......@@ -25,8 +25,8 @@ void gemm<platform::CPUPlace, float>(const CBLAS_TRANSPOSE transA,
const float alpha, const float* A,
const float* B, const float beta, float* C,
platform::DeviceContext* context) {
int lda = K;
int ldb = N;
int lda = (transA == CblasNoTrans) ? K : M;
int ldb = (transB == CblasNoTrans) ? N : K;
int ldc = N;
cblas_sgemm(CblasRowMajor, transA, transB, M, N, K, alpha, A, lda, B, ldb,
beta, C, ldc);
......@@ -40,8 +40,8 @@ void gemm<platform::CPUPlace, double>(const CBLAS_TRANSPOSE transA,
const double* B, const double beta,
double* C,
platform::DeviceContext* context) {
int lda = K;
int ldb = N;
int lda = (transA == CblasNoTrans) ? K : M;
int ldb = (transB == CblasNoTrans) ? N : K;
int ldc = N;
cblas_dgemm(CblasRowMajor, transA, transB, M, N, K, alpha, A, lda, B, ldb,
beta, C, ldc);
......
......@@ -18,6 +18,8 @@
namespace paddle {
namespace operators {
using framework::Tensor;
class MulOp : public framework::OperatorWithKernel {
public:
using framework::OperatorWithKernel::OperatorWithKernel;
......@@ -59,10 +61,23 @@ class MulOpGrad : public framework::OperatorWithKernel {
using framework::OperatorWithKernel::OperatorWithKernel;
protected:
void InferShape(const framework::InferShapeContext &ctx) const override {}
std::string DebugString() const override {
LOG(INFO) << "MulGrad";
return "";
void InferShape(const framework::InferShapeContext &ctx) const override {
PADDLE_ENFORCE_NOT_NULL(ctx.InputVar("X"), "Input(X) should not be null");
PADDLE_ENFORCE_NOT_NULL(ctx.InputVar("Y"), "Input(Y) should not be null");
PADDLE_ENFORCE_NOT_NULL(ctx.InputVar(framework::GradVarName("Out")),
"Input(Out@GRAD) should not be null");
auto x_dims = ctx.Input<Tensor>("X")->dims();
auto y_dims = ctx.Input<Tensor>("Y")->dims();
auto out_dims = ctx.Input<Tensor>(framework::GradVarName("Out"))->dims();
auto *x_grad = ctx.Output<Tensor>(framework::GradVarName("X"));
auto *y_grad = ctx.Output<Tensor>(framework::GradVarName("Y"));
PADDLE_ENFORCE(x_dims[0] == out_dims[0],
"Out@GRAD M X N must equal to X dims 0, M ");
PADDLE_ENFORCE(y_dims[1] == out_dims[1],
"Out@GRAD M X N must equal to Y dims 1, N ");
x_grad->Resize(x_dims);
y_grad->Resize(y_dims);
}
};
......@@ -72,3 +87,5 @@ class MulOpGrad : public framework::OperatorWithKernel {
namespace ops = paddle::operators;
REGISTER_OP(mul, ops::MulOp, ops::MulOpMaker, mul_grad, ops::MulOpGrad);
REGISTER_OP_CPU_KERNEL(mul, ops::MulKernel<paddle::platform::CPUPlace, float>);
REGISTER_OP_CPU_KERNEL(mul_grad,
ops::MulGradKernel<paddle::platform::CPUPlace, float>);
......@@ -17,3 +17,5 @@
namespace ops = paddle::operators;
REGISTER_OP_GPU_KERNEL(mul, ops::MulKernel<paddle::platform::GPUPlace, float>);
REGISTER_OP_GPU_KERNEL(mul_grad,
ops::MulGradKernel<paddle::platform::GPUPlace, float>);
......@@ -31,18 +31,34 @@ template <typename Place, typename T>
class MulKernel : public framework::OpKernel {
public:
void Compute(const framework::ExecutionContext& context) const override {
Eigen::array<Eigen::IndexPair<Eigen::DenseIndex>, 1> dim_pair = {
{Eigen::IndexPair<Eigen::DenseIndex>(1, 0)}};
auto* input0 = context.Input<Tensor>("X");
auto* input1 = context.Input<Tensor>("Y");
auto* output = context.Output<Tensor>("Out");
output->mutable_data<T>(context.GetPlace());
auto X = EigenMatrix<T>::From(*input0);
auto Y = EigenMatrix<T>::From(*input1);
auto Z = EigenMatrix<T>::From(*output);
auto& place = context.GetEigenDevice<Place>();
Z.device(place) = X.contract(Y, dim_pair);
auto* X = context.Input<Tensor>("X");
auto* Y = context.Input<Tensor>("Y");
auto* Z = context.Output<Tensor>("Out");
Z->mutable_data<T>(context.GetPlace());
auto* device_context =
const_cast<platform::DeviceContext*>(context.device_context_);
math::matmul<Place, T>(*X, false, *Y, false, 1, Z, 0, device_context);
}
};
template <typename Place, typename T>
class MulGradKernel : public framework::OpKernel {
public:
void Compute(const framework::ExecutionContext& ctx) const override {
auto* X = ctx.Input<Tensor>("X");
auto* Y = ctx.Input<Tensor>("Y");
auto* dOut = ctx.Input<Tensor>(framework::GradVarName("Out"));
auto* dX = ctx.Output<Tensor>(framework::GradVarName("X"));
auto* dY = ctx.Output<Tensor>(framework::GradVarName("Y"));
dX->mutable_data<T>(ctx.GetPlace());
dY->mutable_data<T>(ctx.GetPlace());
auto* device_context =
const_cast<platform::DeviceContext*>(ctx.device_context_);
// dX = dOut * Y'. dX: M x K, dOut : M x N, Y : K x N
math::matmul<Place, T>(*dOut, false, *Y, true, 1, dX, 0, device_context);
// dY = X' * dOut. dY: K x N, dOut : M x N, X : M x K
math::matmul<Place, T>(*X, true, *dOut, false, 1, dY, 0, device_context);
}
};
......
......@@ -17,7 +17,9 @@
namespace paddle {
namespace operators {
class RowWiseAddOp : public framework::OperatorWithKernel {
using framework::Tensor;
class RowwiseAddOp : public framework::OperatorWithKernel {
public:
using framework::OperatorWithKernel::OperatorWithKernel;
......@@ -34,9 +36,9 @@ class RowWiseAddOp : public framework::OperatorWithKernel {
}
};
class RowWiseAddOpMaker : public framework::OpProtoAndCheckerMaker {
class RowwiseAddOpMaker : public framework::OpProtoAndCheckerMaker {
public:
RowWiseAddOpMaker(framework::OpProto *proto,
RowwiseAddOpMaker(framework::OpProto *proto,
framework::OpAttrChecker *op_checker)
: OpProtoAndCheckerMaker(proto, op_checker) {
AddInput("X", "The left input of row-wise add op, must be matrix");
......@@ -49,12 +51,32 @@ for i in xrange(X.shape[0]):
)DOC");
}
};
class RowwiseAddGradOp : public framework::OperatorWithKernel {
public:
using framework::OperatorWithKernel::OperatorWithKernel;
protected:
void InferShape(const framework::InferShapeContext &ctx) const override {
PADDLE_ENFORCE_NOT_NULL(ctx.InputVar("X"), "X should not be null");
PADDLE_ENFORCE_NOT_NULL(ctx.InputVar("b"), "b should not be null");
PADDLE_ENFORCE_NOT_NULL(ctx.InputVar(framework::GradVarName("Out")),
"Input(Out@GRAD) should not be null");
auto dims0 = ctx.Input<Tensor>("X")->dims();
auto dims1 = ctx.Input<Tensor>("b")->dims();
PADDLE_ENFORCE_EQ(1, dims1.size(), "b dims should be 1")
ctx.Output<Tensor>(framework::GradVarName("X"))->Resize(dims0);
ctx.Output<Tensor>(framework::GradVarName("b"))->Resize(dims1);
}
};
} // namespace operators
} // namespace paddle
namespace ops = paddle::operators;
REGISTER_OP_WITHOUT_GRADIENT(rowwise_add, ops::RowWiseAddOp,
ops::RowWiseAddOpMaker);
REGISTER_OP(rowwise_add, ops::RowwiseAddOp, ops::RowwiseAddOpMaker,
rowwise_add_grad, ops::RowwiseAddGradOp);
REGISTER_OP_CPU_KERNEL(
rowwise_add, ops::RowwiseAddKernel<paddle::platform::CPUPlace, float>);
REGISTER_OP_CPU_KERNEL(
rowwise_add, ops::RowWiseAddKernel<paddle::platform::CPUPlace, float>);
rowwise_add_grad,
ops::RowwiseAddGradKernel<paddle::platform::CPUPlace, float>);
......@@ -17,4 +17,4 @@
namespace ops = paddle::operators;
REGISTER_OP_GPU_KERNEL(
rowwise_add, ops::RowWiseAddKernel<paddle::platform::GPUPlace, float>);
rowwise_add, ops::RowwiseAddKernel<paddle::platform::GPUPlace, float>);
......@@ -28,7 +28,7 @@ template <typename T, int MajorType = Eigen::RowMajor,
using EigenMatrix = framework::EigenMatrix<T, MajorType, IndexType>;
template <typename Place, typename T>
class RowWiseAddKernel : public framework::OpKernel {
class RowwiseAddKernel : public framework::OpKernel {
public:
void Compute(const framework::ExecutionContext& context) const override {
auto out = context.Output<Tensor>("Out");
......@@ -47,5 +47,25 @@ class RowWiseAddKernel : public framework::OpKernel {
}
};
template <typename Place, typename T>
class RowwiseAddGradKernel : public framework::OpKernel {
public:
void Compute(const framework::ExecutionContext& context) const override {
auto* dOut = context.Input<Tensor>(framework::GradVarName("Out"));
auto* dX = context.Output<Tensor>(framework::GradVarName("X"));
auto* db = context.Output<Tensor>(framework::GradVarName("b"));
dX->mutable_data<T>(context.GetPlace());
db->mutable_data<T>(context.GetPlace());
auto OutGrad = EigenMatrix<T>::From(*dOut);
auto place = context.GetEigenDevice<Place>();
EigenMatrix<T>::From(*dX).device(place) = OutGrad;
// https://eigen.tuxfamily.org/dox/unsupported/TensorBase_8h_source.html
// colwise add
Eigen::array<int, 1> dims{{1}}; /* dimension to reduce */
EigenVector<T>::Flatten(*db).device(place) = OutGrad.sum(dims);
}
};
} // namespace operators
} // namespace paddle
cc_library(cpu_info SRCS cpu_info.cc DEPS gflags glog)
cc_test(cpu_info_test SRCS cpu_info_test.cc DEPS cpu_info)
nv_library(gpu_info SRCS gpu_info.cc DEPS gflags)
nv_library(gpu_info SRCS gpu_info.cc DEPS gflags glog)
cc_library(place SRCS place.cc)
cc_test(place_test SRCS place_test.cc DEPS place glog gflags)
......@@ -9,6 +9,7 @@ cc_test(place_test SRCS place_test.cc DEPS place glog gflags)
add_subdirectory(dynload)
cc_test(enforce_test SRCS enforce_test.cc DEPS stringpiece)
cc_test(environment_test SRCS environment_test.cc DEPS stringpiece)
IF(WITH_GPU)
set(GPU_CTX_DEPS dynload_cuda dynamic_loader)
......
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License. */
#pragma once
#include <stdlib.h>
#include <unistd.h>
#include <vector>
#include "paddle/platform/enforce.h"
#include "paddle/string/piece.h"
extern char** environ; // for environment variables
namespace paddle {
namespace platform {
inline void SetEnvVariable(const std::string& name, const std::string& value) {
PADDLE_ENFORCE_NE(setenv(name.c_str(), value.c_str(), 1), -1,
"Failed to set environment variable %s=%s", name, value);
}
inline void UnsetEnvVariable(const std::string& name) {
PADDLE_ENFORCE_NE(unsetenv(name.c_str()), -1,
"Failed to unset environment variable %s", name);
}
inline bool IsEnvVarDefined(const std::string& name) {
return std::getenv(name.c_str()) != nullptr;
}
inline std::string GetEnvValue(const std::string& name) {
PADDLE_ENFORCE(IsEnvVarDefined(name),
"Tried to access undefined environment variable %s", name);
return std::getenv(name.c_str());
}
inline std::vector<std::string> GetAllEnvVariables() {
std::vector<std::string> vars;
for (auto var = environ; *var != nullptr; ++var) {
auto tail = string::Index(*var, "=");
auto name = string::SubStr(*var, 0, tail).ToString();
vars.push_back(name);
}
return vars;
}
} // namespace platform
} // namespace paddle
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License. */
#include "paddle/platform/environment.h"
#include "glog/logging.h"
#include "gtest/gtest.h"
TEST(ENVIRONMENT, ACCESS) {
namespace platform = paddle::platform;
namespace string = paddle::string;
platform::SetEnvVariable("PADDLE_USE_ENV", "TRUE");
EXPECT_TRUE(platform::IsEnvVarDefined("PADDLE_USE_ENV"));
EXPECT_EQ(platform::GetEnvValue("PADDLE_USE_ENV"), "TRUE");
platform::UnsetEnvVariable("PADDLE_USE_ENV");
EXPECT_FALSE(platform::IsEnvVarDefined("PADDLE_USE_ENV"));
platform::SetEnvVariable("PADDLE_USE_ENV1", "Hello ");
platform::SetEnvVariable("PADDLE_USE_ENV2", "World, ");
platform::SetEnvVariable("PADDLE_USE_ENV3", "PaddlePaddle!");
std::string env_info;
auto vars = platform::GetAllEnvVariables();
for_each(vars.begin(), vars.end(), [&](const std::string& var) {
env_info += platform::GetEnvValue(var);
});
EXPECT_TRUE(string::Contains(env_info, "Hello World, PaddlePaddle!"));
platform::UnsetEnvVariable("PADDLE_USE_ENV1");
platform::UnsetEnvVariable("PADDLE_USE_ENV2");
platform::UnsetEnvVariable("PADDLE_USE_ENV3");
env_info.clear();
vars = platform::GetAllEnvVariables();
for_each(vars.begin(), vars.end(), [&](const std::string& var) {
env_info += platform::GetEnvValue(var);
});
EXPECT_FALSE(string::Contains(env_info, "Hello World, PaddlePaddle!"));
EXPECT_FALSE(platform::IsEnvVarDefined("PADDLE_USE_ENV1"));
EXPECT_FALSE(platform::IsEnvVarDefined("PADDLE_USE_ENV2"));
EXPECT_FALSE(platform::IsEnvVarDefined("PADDLE_USE_ENV3"));
}
......@@ -13,8 +13,11 @@ See the License for the specific language governing permissions and
limitations under the License. */
#include "paddle/platform/gpu_info.h"
#include "gflags/gflags.h"
#include "paddle/platform/enforce.h"
#include "paddle/platform/environment.h"
DEFINE_double(fraction_of_gpu_memory_to_use, 0.95,
"Default use 95% of GPU memory for PaddlePaddle,"
......@@ -70,6 +73,13 @@ size_t GpuMaxChunkSize() {
GpuMemoryUsage(available, total);
if (IsEnvVarDefined(kEnvFractionGpuMemoryToUse)) {
auto val = std::stod(GetEnvValue(kEnvFractionGpuMemoryToUse));
PADDLE_ENFORCE_GT(val, 0.0);
PADDLE_ENFORCE_LE(val, 1.0);
FLAGS_fraction_of_gpu_memory_to_use = val;
}
// Reserving the rest memory for page tables, etc.
size_t reserving = (1 - FLAGS_fraction_of_gpu_memory_to_use) * total;
......
......@@ -18,10 +18,15 @@ limitations under the License. */
#include <cuda_runtime.h>
#include <stddef.h>
#include <string>
namespace paddle {
namespace platform {
//! Environment variable: fraction of GPU memory to use on each device.
const std::string kEnvFractionGpuMemoryToUse =
"PADDLE_FRACTION_GPU_MEMORY_TO_USE";
//! Get the total number of GPU devices in system.
int GetDeviceCount();
......
import unittest
from op_test_util import OpTestMeta
import numpy as np
from gradient_checker import GradientChecker, create_op
from op_test_util import OpTestMeta
class TestMulOp(unittest.TestCase):
......@@ -15,5 +16,19 @@ class TestMulOp(unittest.TestCase):
self.outputs = {'Out': np.dot(self.inputs['X'], self.inputs['Y'])}
class MulGradOpTest(GradientChecker):
def test_mul(self):
op = create_op("mul")
inputs = {
'X': np.random.random((32, 84)).astype("float32"),
'Y': np.random.random((84, 100)).astype("float32")
}
# mul op will enlarge the relative error
self.check_grad(
op, inputs, set(["X", "Y"]), "Out", max_relative_error=0.5)
# TODO(dzh,qijun) : mulgrad test case need transpose feature of blas library
if __name__ == '__main__':
unittest.main()
import unittest
from op_test_util import OpTestMeta
import numpy as np
from op_test_util import OpTestMeta
from gradient_checker import GradientChecker, create_op
class TestRowwiseAddOp(unittest.TestCase):
......@@ -15,5 +16,15 @@ class TestRowwiseAddOp(unittest.TestCase):
self.outputs = {'Out': np.add(self.inputs['X'], self.inputs['b'])}
class RowwiseAddGradOpTest(GradientChecker):
def test_rowwise_add(self):
op = create_op("rowwise_add")
inputs = {
"X": np.random.uniform(0.1, 1, [10, 10]).astype("float32"),
"b": np.random.uniform(0.1, 1, [10]).astype("float32")
}
self.check_grad(op, inputs, set(["X", "b"]), "Out")
if __name__ == '__main__':
unittest.main()
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册
反馈
建议
客服 返回
顶部