Nigel
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

!16629 Binary_cross_entropy op supports optional weight input 

From: @zuochuanyong
Reviewed-by: @zhoufeng54,@liangchenghui
Signed-off-by: @liangchenghui
This commit is contained in:
mindspore-ci-bot 2021-05-25 09:26:00 +08:00 committed by Gitee
parent f64bc04f95 fc06d0562c
commit e1803a0938
11 changed files with 193 additions and 37 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
mindspore/ccsrc/backend/kernel_compiler/cpu/binary_cross_entropy_cpu_kernel.cc
View File

@ -17,6 +17,8 @@
namespace mindspore {
namespace kernel {
constexpr size_t kBceInputNumWithWeight = 3;
template <typename T>
void BinaryCrossEntropyCpuKernel::LaunchToScalar(const int &input_size, const int &reduction, T *loss, T *tmp_loss) {
if (input_size % 2 == 1) {
@ -44,24 +46,37 @@ void BinaryCrossEntropyCpuKernel::Launchkernel(const std::vector<AddressPtr> &in
const std::vector<AddressPtr> &outputs) {
T *input_x = reinterpret_cast<T *>(inputs[0]->addr);
T *input_y = reinterpret_cast<T *>(inputs[1]->addr);
T *weight = reinterpret_cast<T *>(inputs[2]->addr);
T *weight = nullptr;
if (weight_defined_) {
weight = reinterpret_cast<T *>(inputs[2]->addr);
}
T *loss = reinterpret_cast<T *>(outputs[0]->addr);
std::vector<T> tmp_loss(input_size_);
T epsilon = static_cast<T>(1e-12);
T one = static_cast<T>(1);
if (reduction_ == 0) {
if (reduction_ == 0 && weight_defined_) {
for (size_t i = 0; i < input_size_; i++) {
T value =
-weight[i] * (input_y[i] * log(input_x[i] + epsilon) + (one - input_y[i]) * log(one - input_x[i] + epsilon));
loss[i] = value;
}
} else {
} else if (reduction_ == 0 && (!weight_defined_)) {
for (size_t i = 0; i < input_size_; i++) {
T value = -(input_y[i] * log(input_x[i] + epsilon) + (one - input_y[i]) * log(one - input_x[i] + epsilon));
loss[i] = value;
}
} else if ((reduction_ != 0) && weight_defined_) {
for (size_t i = 0; i < input_size_; i++) {
T value =
-weight[i] * (input_y[i] * log(input_x[i] + epsilon) + (one - input_y[i]) * log(one - input_x[i] + epsilon));
tmp_loss[i] = value;
}
} else {
for (size_t i = 0; i < input_size_; i++) {
T value = -(input_y[i] * log(input_x[i] + epsilon) + (one - input_y[i]) * log(one - input_x[i] + epsilon));
tmp_loss[i] = value;
}
}
if (reduction_ != 0) {
@ -93,7 +108,8 @@ void BinaryCrossEntropyCpuKernel::InitKernel(const CNodePtr &kernel_node) {
} else if (reduction == "sum") {
reduction_ = 2;
}
size_t input_num = AnfAlgo::GetInputTensorNum(kernel_node);
weight_defined_ = (input_num == kBceInputNumWithWeight);
dtype_ = AnfAlgo::GetInputDeviceDataType(kernel_node, 0);
}
} // namespace kernel

11
mindspore/ccsrc/backend/kernel_compiler/cpu/binary_cross_entropy_cpu_kernel.h
View File

@ -25,7 +25,7 @@ namespace mindspore {
namespace kernel {
class BinaryCrossEntropyCpuKernel : public CPUKernel {
public:
BinaryCrossEntropyCpuKernel() : input_size_(1), reduction_(1) {}
BinaryCrossEntropyCpuKernel() : input_size_(1), reduction_(1), weight_defined_(false) {}
~BinaryCrossEntropyCpuKernel() override = default;
void InitKernel(const CNodePtr &kernel_node) override;
@ -42,6 +42,7 @@ class BinaryCrossEntropyCpuKernel : public CPUKernel {
TypeId dtype_{kTypeUnknown};
size_t input_size_;
int reduction_;
bool weight_defined_; // true: there are 3 inputs, false: there are 2 inputs(no [weight])
};
MS_REG_CPU_KERNEL(BinaryCrossEntropy,
KernelAttr()
@ -57,6 +58,14 @@ MS_REG_CPU_KERNEL(BinaryCrossEntropy,
.AddInputAttr(kNumberTypeFloat32)
.AddOutputAttr(kNumberTypeFloat32),
BinaryCrossEntropyCpuKernel);
MS_REG_CPU_KERNEL(
BinaryCrossEntropy,
KernelAttr().AddInputAttr(kNumberTypeFloat16).AddInputAttr(kNumberTypeFloat16).AddOutputAttr(kNumberTypeFloat16),
BinaryCrossEntropyCpuKernel);
MS_REG_CPU_KERNEL(
BinaryCrossEntropy,
KernelAttr().AddInputAttr(kNumberTypeFloat32).AddInputAttr(kNumberTypeFloat32).AddOutputAttr(kNumberTypeFloat32),
BinaryCrossEntropyCpuKernel);
} // namespace kernel
} // namespace mindspore
#endif // MINDSPORE_CCSRC_BACKEND_KERNEL_COMPILER_CPU_NN_BINARY_CROSS_ENTROPY_KERNEL_H

42
mindspore/ccsrc/backend/kernel_compiler/cpu/binary_cross_entropy_grad_kernel.cc
View File

@ -17,32 +17,54 @@
namespace mindspore {
namespace kernel {
constexpr size_t kBceGradInputNumWithWeight = 4;
template <typename T>
void BinaryCrossEntropyGradCpuKernel::Launchkernel(const std::vector<AddressPtr> &inputs,
const std::vector<AddressPtr> &outputs) {
T *input_x = reinterpret_cast<T *>(inputs[0]->addr);
T *input_y = reinterpret_cast<T *>(inputs[1]->addr);
T *dloss = reinterpret_cast<T *>(inputs[2]->addr);
T *weight = reinterpret_cast<T *>(inputs[3]->addr);
T *weight = nullptr;
if (weight_defined_) {
weight = reinterpret_cast<T *>(inputs[3]->addr);
}
T *dx = reinterpret_cast<T *>(outputs[0]->addr);
T epsilon = static_cast<T>(1e-12);
T one = static_cast<T>(1);
if (reduction_ == 0) {
for (size_t i = 0; i < input_size_; i++) {
T denominator = ((input_x[i] * (one - input_x[i])) > epsilon) ? (input_x[i] * (one - input_x[i])) : epsilon;
T value = weight[i] * (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss[i];
if (weight_defined_) {
for (size_t i = 0; i < input_size_; i++) {
T denominator = ((input_x[i] * (one - input_x[i])) > epsilon) ? (input_x[i] * (one - input_x[i])) : epsilon;
T value = weight[i] * (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss[i];
}
} else {
for (size_t i = 0; i < input_size_; i++) {
T denominator = ((input_x[i] * (one - input_x[i])) > epsilon) ? (input_x[i] * (one - input_x[i])) : epsilon;
T value = (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss[i];
}
}
} else {
T dloss1 = dloss[0];
if (reduction_ == 1) {
dloss1 = dloss[0] / static_cast<T>(input_size_);
}
for (size_t i = 0; i < input_size_; i++) {
T denominator = ((input_x[i] * (one - input_x[i])) > epsilon) ? (input_x[i] * (one - input_x[i])) : epsilon;
T value = weight[i] * (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss1;
if (weight_defined_) {
for (size_t i = 0; i < input_size_; i++) {
T denominator = ((input_x[i] * (one - input_x[i])) > epsilon) ? (input_x[i] * (one - input_x[i])) : epsilon;
T value = weight[i] * (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss1;
}
} else {
for (size_t i = 0; i < input_size_; i++) {
T denominator = ((input_x[i] * (one - input_x[i])) > epsilon) ? (input_x[i] * (one - input_x[i])) : epsilon;
T value = (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss1;
}
}
}
}
@ -72,6 +94,8 @@ void BinaryCrossEntropyGradCpuKernel::InitKernel(const CNodePtr &kernel_node) {
reduction_ = 2;
}
size_t input_num = AnfAlgo::GetInputTensorNum(kernel_node);
weight_defined_ = (input_num == kBceGradInputNumWithWeight);
dtype_ = AnfAlgo::GetInputDeviceDataType(kernel_node, 0);
}
} // namespace kernel

17
mindspore/ccsrc/backend/kernel_compiler/cpu/binary_cross_entropy_grad_kernel.h
View File

@ -25,7 +25,7 @@ namespace mindspore {
namespace kernel {
class BinaryCrossEntropyGradCpuKernel : public CPUKernel {
public:
BinaryCrossEntropyGradCpuKernel() : input_size_(1), reduction_(1) {}
BinaryCrossEntropyGradCpuKernel() : input_size_(1), reduction_(1), weight_defined_(false) {}
~BinaryCrossEntropyGradCpuKernel() override = default;
void InitKernel(const CNodePtr &kernel_node) override;
@ -39,6 +39,7 @@ class BinaryCrossEntropyGradCpuKernel : public CPUKernel {
TypeId dtype_{kTypeUnknown};
size_t input_size_;
int reduction_;
bool weight_defined_; // true: there are 4 inputs, false: there are 3 inputs(no [weight])
};
MS_REG_CPU_KERNEL(BinaryCrossEntropyGrad,
KernelAttr()
@ -56,6 +57,20 @@ MS_REG_CPU_KERNEL(BinaryCrossEntropyGrad,
.AddInputAttr(kNumberTypeFloat32)
.AddOutputAttr(kNumberTypeFloat32),
BinaryCrossEntropyGradCpuKernel);
MS_REG_CPU_KERNEL(BinaryCrossEntropyGrad,
KernelAttr()
.AddInputAttr(kNumberTypeFloat16)
.AddInputAttr(kNumberTypeFloat16)
.AddInputAttr(kNumberTypeFloat16)
.AddOutputAttr(kNumberTypeFloat16),
BinaryCrossEntropyGradCpuKernel);
MS_REG_CPU_KERNEL(BinaryCrossEntropyGrad,
KernelAttr()
.AddInputAttr(kNumberTypeFloat32)
.AddInputAttr(kNumberTypeFloat32)
.AddInputAttr(kNumberTypeFloat32)
.AddOutputAttr(kNumberTypeFloat32),
BinaryCrossEntropyGradCpuKernel);
} // namespace kernel
} // namespace mindspore
#endif // MINDSPORE_CCSRC_BACKEND_KERNEL_COMPILER_CPU_NN_BINARY_CROSS_ENTROPY_GRAD_KERNEL_H

46
mindspore/ccsrc/backend/kernel_compiler/gpu/cuda_impl/loss_with_reduction_impl.cu
View File

@ -124,18 +124,28 @@ __global__ void BinaryCrossEntropyLossKernel(const int input_size, const int red
const T *input_y, const T *weight, T *loss, T *tmp_loss) {
T epsilon = 1e-12;
T one = static_cast<T>(1);
if (reduction == 0) {
if (reduction == 0 && weight != nullptr) {
for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < input_size; i += blockDim.x * gridDim.x) {
T value =
-weight[i] * (input_y[i] * logT(input_x[i] + epsilon) + (one - input_y[i]) * logT(one - input_x[i] + epsilon));
loss[i] = value;
}
} else {
} else if (reduction == 0 && weight == nullptr) {
for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < input_size; i += blockDim.x * gridDim.x) {
T value = -(input_y[i] * logT(input_x[i] + epsilon) + (one - input_y[i]) * logT(one - input_x[i] + epsilon));
loss[i] = value;
}
} else if (reduction != 0 && weight != nullptr) {
for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < input_size; i += blockDim.x * gridDim.x) {
T value =
-weight[i] * (input_y[i] * logT(input_x[i] + epsilon) + (one - input_y[i]) * logT(one - input_x[i] + epsilon));
tmp_loss[i] = value;
}
} else {
for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < input_size; i += blockDim.x * gridDim.x) {
T value = -(input_y[i] * logT(input_x[i] + epsilon) + (one - input_y[i]) * logT(one - input_x[i] + epsilon));
tmp_loss[i] = value;
}
}
}
@ -165,20 +175,36 @@ __global__ void BinaryCrossEntropyLossGradKernel(const int input_size, const int
T epsilon = 1e-12;
T one = static_cast<T>(1);
if (reduction == 0) {
for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < input_size; i += blockDim.x * gridDim.x) {
T denominator = maxT(input_x[i] * (one - input_x[i]), epsilon);
T value = weight[i] * (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss[i];
if (weight != nullptr) {
for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < input_size; i += blockDim.x * gridDim.x) {
T denominator = maxT(input_x[i] * (one - input_x[i]), epsilon);
T value = weight[i] * (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss[i];
}
} else {
for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < input_size; i += blockDim.x * gridDim.x) {
T denominator = maxT(input_x[i] * (one - input_x[i]), epsilon);
T value = (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss[i];
}
}
} else {
T dloss1 = dloss[0];
if (reduction == 1) {
dloss1 = dloss[0] / castT(dloss[0], input_size);
}
for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < input_size; i += blockDim.x * gridDim.x) {
T denominator = maxT(input_x[i] * (one - input_x[i]), epsilon);
T value = weight[i] * (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss1;
if (weight != nullptr) {
for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < input_size; i += blockDim.x * gridDim.x) {
T denominator = maxT(input_x[i] * (one - input_x[i]), epsilon);
T value = weight[i] * (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss1;
}
} else {
for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < input_size; i += blockDim.x * gridDim.x) {
T denominator = maxT(input_x[i] * (one - input_x[i]), epsilon);
T value = (input_x[i] - input_y[i]) / denominator;
dx[i] = value * dloss1;
}
}
}
}

8
mindspore/ccsrc/backend/kernel_compiler/gpu/nn/binary_cross_entropy_gpu_kernel.cc
View File

@ -31,5 +31,13 @@ MS_REG_GPU_KERNEL_ONE(BinaryCrossEntropy,
.AddInputAttr(kNumberTypeFloat16)
.AddOutputAttr(kNumberTypeFloat16),
BinaryCrossEntropyGpuKernel, half)
MS_REG_GPU_KERNEL_ONE(
BinaryCrossEntropy,
KernelAttr().AddInputAttr(kNumberTypeFloat32).AddInputAttr(kNumberTypeFloat32).AddOutputAttr(kNumberTypeFloat32),
BinaryCrossEntropyGpuKernel, float)
MS_REG_GPU_KERNEL_ONE(
BinaryCrossEntropy,
KernelAttr().AddInputAttr(kNumberTypeFloat16).AddInputAttr(kNumberTypeFloat16).AddOutputAttr(kNumberTypeFloat16),
BinaryCrossEntropyGpuKernel, half)
} // namespace kernel
} // namespace mindspore

14
mindspore/ccsrc/backend/kernel_compiler/gpu/nn/binary_cross_entropy_gpu_kernel.h
View File

@ -28,7 +28,7 @@ namespace kernel {
template <typename T>
class BinaryCrossEntropyGpuKernel : public GpuKernel {
public:
BinaryCrossEntropyGpuKernel() : input_size_(1), reduction_(1) {}
BinaryCrossEntropyGpuKernel() : weight_defined_(false), input_size_(1), reduction_(1) {}
~BinaryCrossEntropyGpuKernel() override = default;
const std::vector<size_t> &GetInputSizeList() const override { return input_size_list_; }
const std::vector<size_t> &GetOutputSizeList() const override { return output_size_list_; }
@ -37,7 +37,10 @@ class BinaryCrossEntropyGpuKernel : public GpuKernel {
const std::vector<AddressPtr> &outputs, void *stream_ptr) override {
T *input_x = GetDeviceAddress<T>(inputs, 0);
T *input_y = GetDeviceAddress<T>(inputs, 1);
T *weight = GetDeviceAddress<T>(inputs, 2);
T *weight = nullptr;
if (weight_defined_) {
weight = GetDeviceAddress<T>(inputs, 2);
}
T *loss = GetDeviceAddress<T>(outputs, 0);
T *tmp_loss = GetDeviceAddress<T>(workspace, 0);
if (input_size_ > 0) {
@ -49,6 +52,8 @@ class BinaryCrossEntropyGpuKernel : public GpuKernel {
bool Init(const CNodePtr &kernel_node) override {
auto input_shape = AnfAlgo::GetPrevNodeOutputInferShape(kernel_node, 0);
size_t input_num = AnfAlgo::GetInputTensorNum(kernel_node);
weight_defined_ = (input_num == 3);
for (size_t i = 0; i < input_shape.size(); i++) {
input_size_ *= input_shape[i];
}
@ -70,7 +75,9 @@ class BinaryCrossEntropyGpuKernel : public GpuKernel {
void InitSizeLists() override {
input_size_list_.push_back(input_size_ * sizeof(T));
input_size_list_.push_back(input_size_ * sizeof(T));
input_size_list_.push_back(input_size_ * sizeof(T));
if (weight_defined_) {
input_size_list_.push_back(input_size_ * sizeof(T));
}
if (reduction_ == 0) {
output_size_list_.push_back(input_size_ * sizeof(T));
} else {
@ -80,6 +87,7 @@ class BinaryCrossEntropyGpuKernel : public GpuKernel {
}
private:
bool weight_defined_; // true: there are 3 inputs, false: there are 2 inputs(no [weight])
size_t input_size_;
int reduction_;
size_t workspace_size_;

14
mindspore/ccsrc/backend/kernel_compiler/gpu/nn/binary_cross_entropy_grad_kernel.cc
View File

@ -34,5 +34,19 @@ MS_REG_GPU_KERNEL_ONE(BinaryCrossEntropyGrad,
.AddInputAttr(kNumberTypeFloat16)
.AddOutputAttr(kNumberTypeFloat16),
BinaryCrossEntropyGradGpuKernel, half)
MS_REG_GPU_KERNEL_ONE(BinaryCrossEntropyGrad,
KernelAttr()
.AddInputAttr(kNumberTypeFloat32)
.AddInputAttr(kNumberTypeFloat32)
.AddInputAttr(kNumberTypeFloat32)
.AddOutputAttr(kNumberTypeFloat32),
BinaryCrossEntropyGradGpuKernel, float)
MS_REG_GPU_KERNEL_ONE(BinaryCrossEntropyGrad,
KernelAttr()
.AddInputAttr(kNumberTypeFloat16)
.AddInputAttr(kNumberTypeFloat16)
.AddInputAttr(kNumberTypeFloat16)
.AddOutputAttr(kNumberTypeFloat16),
BinaryCrossEntropyGradGpuKernel, half)
} // namespace kernel
} // namespace mindspore

15
mindspore/ccsrc/backend/kernel_compiler/gpu/nn/binary_cross_entropy_grad_kernel.h
View File

@ -28,7 +28,7 @@ namespace kernel {
template <typename T>
class BinaryCrossEntropyGradGpuKernel : public GpuKernel {
public:
BinaryCrossEntropyGradGpuKernel() : input_size_(1), reduction_(1) {}
BinaryCrossEntropyGradGpuKernel() : input_size_(1), reduction_(1), weight_defined_(false) {}
~BinaryCrossEntropyGradGpuKernel() override = default;
const std::vector<size_t> &GetInputSizeList() const override { return input_size_list_; }
@ -40,7 +40,10 @@ class BinaryCrossEntropyGradGpuKernel : public GpuKernel {
T *input_x = GetDeviceAddress<T>(inputs, 0);
T *input_y = GetDeviceAddress<T>(inputs, 1);
T *dloss = GetDeviceAddress<T>(inputs, 2);
T *weight = GetDeviceAddress<T>(inputs, 3);
T *weight = nullptr;
if (weight_defined_) {
weight = GetDeviceAddress<T>(inputs, 3);
}
T *dx = GetDeviceAddress<T>(outputs, 0);
if (input_size_ > 0) {
BinaryCrossEntropyLossGrad(input_size_, reduction_, input_x, input_y, weight, dloss, dx,
@ -51,6 +54,8 @@ class BinaryCrossEntropyGradGpuKernel : public GpuKernel {
bool Init(const CNodePtr &kernel_node) override {
auto input_shape = AnfAlgo::GetPrevNodeOutputInferShape(kernel_node, 0);
size_t input_num = AnfAlgo::GetInputTensorNum(kernel_node);
weight_defined_ = (input_num == 4);
for (size_t i = 0; i < input_shape.size(); i++) {
input_size_ *= input_shape[i];
}
@ -73,14 +78,16 @@ class BinaryCrossEntropyGradGpuKernel : public GpuKernel {
} else {
input_size_list_.push_back(sizeof(T));
}
input_size_list_.push_back(input_size_ * sizeof(T));
if (weight_defined_) {
input_size_list_.push_back(input_size_ * sizeof(T));
}
output_size_list_.push_back(input_size_ * sizeof(T));
}
private:
size_t input_size_;
int reduction_;
bool weight_defined_; // true: there are 4 inputs, false: there are 3 inputs(no [weight])
std::vector<size_t> input_size_list_;
std::vector<size_t> output_size_list_;
std::vector<size_t> workspace_size_list_;

22
tests/st/ops/cpu/test_binary_cross_entropy_op.py
View File

@ -24,12 +24,13 @@ from mindspore.ops import operations as P
context.set_context(mode=context.GRAPH_MODE, device_target="CPU")
class Net(nn.Cell):
def __init__(self, reduction="none"):
super(Net, self).__init__()
self.BinaryCrossEntropy = P.BinaryCrossEntropy(reduction)
def construct(self, x, y, weight):
def construct(self, x, y, weight=None):
return self.BinaryCrossEntropy(x, y, weight)
@ -50,6 +51,7 @@ def test_binary_cross_entropy_loss():
0.03405444, 0.23934692]
assert np.allclose(loss.asnumpy(), expect)
def test_binary_cross_entropy_loss_mean():
np.random.seed(42)
prediction = np.random.rand(20).astype(np.float32)
@ -61,6 +63,7 @@ def test_binary_cross_entropy_loss_mean():
expect = [0.7447324991226196]
assert loss.asnumpy() == expect
def test_binary_cross_entropy_loss_sum():
np.random.seed(42)
prediction = np.random.rand(20).astype(np.float32)
@ -72,6 +75,18 @@ def test_binary_cross_entropy_loss_sum():
expect = [14.894649505615234]
assert loss.asnumpy() == expect
def test_binary_cross_entropy_loss_sum_without_weight():
np.random.seed(42)
prediction = np.random.rand(20).astype(np.float32)
target = np.random.rand(20).astype(np.float32)
reduction = "sum"
net = Net(reduction)
loss = net(Tensor(prediction), Tensor(target))
expect = [25.48195216753522]
assert np.allclose(loss.asnumpy(), expect)
def test_binary_cross_entropy_loss_16():
np.random.seed(42)
prediction = np.random.rand(20).astype(np.float16)
@ -86,6 +101,7 @@ def test_binary_cross_entropy_loss_16():
0.0340576, 0.239258]
assert np.allclose(loss.asnumpy(), expect)
def test_binary_cross_entropy_loss_mean_16():
np.random.seed(42)
prediction = np.random.rand(20).astype(np.float16)
@ -97,6 +113,7 @@ def test_binary_cross_entropy_loss_mean_16():
expect = [0.74462890625]
assert loss.asnumpy() == expect
def test_binary_cross_entropy_loss_sum_16():
np.random.seed(42)
prediction = np.random.rand(20).astype(np.float16)
@ -108,13 +125,14 @@ def test_binary_cross_entropy_loss_sum_16():
expect = [14.890625]
assert loss.asnumpy() == expect
class Grad(nn.Cell):
def __init__(self, network):
super(Grad, self).__init__()
self.grad = C.GradOperation(get_all=True, sens_param=True)
self.network = network
def construct(self, x1, x2, sens, weight):
def construct(self, x1, x2, sens, weight=None):
gout = self.grad(self.network)(x1, x2, sens, weight)
return gout

17
tests/st/ops/gpu/test_binary_cross_entropy_op.py
View File

@ -28,9 +28,9 @@ context.set_context(mode=context.GRAPH_MODE, device_target="GPU")
class Net(nn.Cell):
def __init__(self, reduction="none"):
super(Net, self).__init__()
self.BinaryCrossEntropy = P.BinaryCrossEntropy("none")
self.BinaryCrossEntropy = P.BinaryCrossEntropy(reduction)
def construct(self, x, y, weight):
def construct(self, x, y, weight=None):
return self.BinaryCrossEntropy(x, y, weight)
@ -51,13 +51,24 @@ def test_binary_cross_entropy_loss():
assert np.allclose(loss.asnumpy(), expect)
def test_binary_cross_entropy_loss_sum_without_weight():
np.random.seed(42)
prediction = np.random.rand(20).astype(np.float32)
target = np.random.rand(20).astype(np.float32)
reduction = "sum"
net = Net(reduction)
loss = net(Tensor(prediction), Tensor(target))
expect = [25.48195216753522]
assert np.allclose(loss.asnumpy(), expect)
class Grad(nn.Cell):
def __init__(self, network):
super(Grad, self).__init__()
self.grad = C.GradOperation(get_all=True, sens_param=True)
self.network = network
def construct(self, x1, x2, sens, weight):
def construct(self, x1, x2, sens, weight=None):
gout = self.grad(self.network)(x1, x2, sens, weight)
return gout