代码
注册 登录
简体中文
简体中文 English
openoker
/
DeepSpeed
1
0
收藏 0
文件 工单管理 0 Wiki
分支: qanthony/comms-bench
分支列表 标签列表
Ramya Ramineni 7bcb4fabeb Enable CG headers on ROCm (#1821) 2 年之前
..
StopWatch.h 734d8991c8 Transformer kernel release (#242) 4 年之前
Timer.h a10e4811fe force set lf instead of crlf (https://github.com/pre-commit/pre-commit-hooks#mixed-line-ending) (#1598) 2 年之前
compat.h 8abdaee243 Add cpu adagrad (#1358) 3 年之前
context.h ee1ffe2e88 CPU-Adam fix for scalar mode (#735) 3 年之前
cpu_adagrad.h 74493b2bee support CPU Adam and Adagrad on Windows with SDK 10.0.22000 (#1634) 2 年之前
cpu_adam.h 74493b2bee support CPU Adam and Adagrad on Windows with SDK 10.0.22000 (#1634) 2 年之前
cublas_wrappers.h c3c8d5dd93 AMD support (#1430) 2 年之前
custom_cuda_layers.h c3c8d5dd93 AMD support (#1430) 2 年之前
dropout.h a10e4811fe force set lf instead of crlf (https://github.com/pre-commit/pre-commit-hooks#mixed-line-ending) (#1598) 2 年之前
ds_transformer_cuda.h bc7778ea5b Fix the workspace allocation for the transformer kernel (#1397) 3 年之前
feed_forward.h c3c8d5dd93 AMD support (#1430) 2 年之前
gelu.h a10e4811fe force set lf instead of crlf (https://github.com/pre-commit/pre-commit-hooks#mixed-line-ending) (#1598) 2 年之前
gemm_test.h c3c8d5dd93 AMD support (#1430) 2 年之前
general_kernels.h c3c8d5dd93 AMD support (#1430) 2 年之前
normalize_layer.h a10e4811fe force set lf instead of crlf (https://github.com/pre-commit/pre-commit-hooks#mixed-line-ending) (#1598) 2 年之前
quantizer.h ed3de0c21b Quantization + inference release (#1091) 3 年之前
simd.h 259936a76c Fix cpu-adam AVX performance (#1637) 2 年之前
softmax.h a10e4811fe force set lf instead of crlf (https://github.com/pre-commit/pre-commit-hooks#mixed-line-ending) (#1598) 2 年之前
strided_batch_gemm.h c3c8d5dd93 AMD support (#1430) 2 年之前
type_shim.h 648f7bfa50 Bfloat16 zero2 (#1398) 3 年之前