You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
ColossalAI/extensions/pybind/optimizer/optimizer.cpp

50 lines
2.5 KiB

// modified from
// https://github.com/NVIDIA/apex/blob/master/csrc/multi_tensor_adam.cu
3 years ago
#include <torch/extension.h>
void multi_tensor_scale_cuda(int chunk_size, at::Tensor noop_flag,
std::vector<std::vector<at::Tensor>> tensor_lists,
float scale);
3 years ago
void multi_tensor_sgd_cuda(int chunk_size, at::Tensor noop_flag,
std::vector<std::vector<at::Tensor>> tensor_lists,
float wd, float momentum, float dampening, float lr,
bool nesterov, bool first_run,
bool wd_after_momentum, float scale);
3 years ago
void multi_tensor_adam_cuda(int chunk_size, at::Tensor noop_flag,
std::vector<std::vector<at::Tensor>> tensor_lists,
const float lr, const float beta1,
const float beta2, const float epsilon,
const int step, const int mode,
const int bias_correction, const float weight_decay,
const float div_scale);
3 years ago
void multi_tensor_lamb_cuda(int chunk_size, at::Tensor noop_flag,
std::vector<std::vector<at::Tensor>> tensor_lists,
const float lr, const float beta1,
const float beta2, const float epsilon,
const int step, const int bias_correction,
const float weight_decay, const int grad_averaging,
const int mode, at::Tensor global_grad_norm,
const float max_grad_norm,
at::optional<bool> use_nvlamb_python);
3 years ago
std::tuple<at::Tensor, at::Tensor> multi_tensor_l2norm_cuda(
int chunk_size, at::Tensor noop_flag,
std::vector<std::vector<at::Tensor>> tensor_lists,
at::optional<bool> per_tensor_python);
3 years ago
PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
m.def("multi_tensor_scale", &multi_tensor_scale_cuda,
3 years ago
"Fused overflow check + scale for a list of contiguous tensors");
m.def("multi_tensor_sgd", &multi_tensor_sgd_cuda,
"Fused SGD optimizer for list of contiguous tensors");
m.def("multi_tensor_adam", &multi_tensor_adam_cuda,
"Compute and apply gradient update to parameters for Adam optimizer");
m.def("multi_tensor_lamb", &multi_tensor_lamb_cuda,
"Computes and apply update for LAMB optimizer");
m.def("multi_tensor_l2norm", &multi_tensor_l2norm_cuda,
"Computes L2 norm for a list of contiguous tensors");
}