OpenI
/
MegEngine
Not watched
1
0
Fork 0
Code
Releases 31 Wiki Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
2266 Commits
28 Branches
25 MB
3465 Download
 
 
 
 
 
 
Tree: 39d77fb55a
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from '39d77fb55a'
${ noResults }
MegEngine/dnn/test/cuda
Megvii Engine Team cbbca5fb10 feat(mge): add softmax op use cudnn api 3 years ago
..
local style(all): reformat c++ code 3 years ago
megcore style(all): reformat c++ code 3 years ago
accuracy_shake.cpp test(dnn): fix compute capability requirement for NCHWX test 3 years ago
adaptive_pooling.cpp feat(dnn): use ref ptr interface for all backends 3 years ago
add_update.cpp style(all): reformat c++ code 3 years ago
argmxx.cpp style(all): reformat c++ code 3 years ago
argsort.cpp style(all): reformat c++ code 3 years ago
batch_conv_bias.cpp style(all): reformat c++ code 3 years ago
batched_matrix_mul.cpp style(all): reformat c++ code 3 years ago
benchmark.cpp style(all): reformat c++ code 3 years ago
benchmark.h style(all): reformat c++ code 3 years ago
bn.cpp feat(dnn): use ref ptr interface for all backends 3 years ago
chanwise_convolution.cpp style(all): reformat c++ code 3 years ago
chanwise_convolution3d.cpp style(all): reformat c++ code 3 years ago
check_non_finite.cpp feat(dnn/check_non_finite): add batch check_non_finite 3 years ago
checksum.cpp feat(dnn): use ref ptr interface for all backends 3 years ago
concat.cpp style(all): reformat c++ code 3 years ago
cond_take.cpp style(all): reformat c++ code 3 years ago
conv_bias.cpp fix(dnn/cuda): fix cudnn_conv algo of conv_bias opr for fp16 add z cases 3 years ago
conv_bias_int8.cpp style(all): reformat c++ code 3 years ago
conv_pooling.cpp style(all): reformat c++ code 3 years ago
conv_test_utils.cpp style(all): reformat c++ code 3 years ago
conv_test_utils.h style(all): reformat c++ code 3 years ago
convolution.cpp style(all): reformat c++ code 3 years ago
convolution3d.cpp style(all): reformat c++ code 3 years ago
correlation.cpp style(all): reformat c++ code 3 years ago
cumsum.cpp style(all): reformat c++ code 3 years ago
cutlass_matmul.cpp style(all): reformat c++ code 3 years ago
cvt_color.cpp style(all): reformat c++ code 3 years ago
dct.cpp style(all): reformat c++ code 3 years ago
deformable_conv.cpp style(all): reformat c++ code 3 years ago
deformable_ps_roi_pooling.cpp style(all): reformat c++ code 3 years ago
dilated_convolution.cpp style(all): reformat c++ code 3 years ago
dot.cpp style(all): reformat c++ code 3 years ago
elemwise.cpp feat(dnn): use ref ptr interface for all backends 3 years ago
elemwise_multi_type.cpp style(all): reformat c++ code 3 years ago
eye.cpp style(all): reformat c++ code 3 years ago
fake_quant.cpp fix(dnn): correct reduce/argmxx/fakequant calculation with nan 3 years ago
fill.cpp style(all): reformat c++ code 3 years ago
fixture.cpp style(all): reformat c++ code 3 years ago
fixture.h style(all): reformat c++ code 3 years ago
flip.cpp style(all): reformat c++ code 3 years ago
gaussian_blur.cpp style(all): reformat c++ code 3 years ago
group_conv.cpp style(all): reformat c++ code 3 years ago
group_conv3d.cpp style(all): reformat c++ code 3 years ago
group_local.cpp style(all): reformat c++ code 3 years ago
images2neibs.cpp style(all): reformat c++ code 3 years ago
indexing_multi_axis_vec.cpp feat(opr): indexing_multi_axis_vec support nd index 3 years ago
indexing_one_hot.cpp style(all): reformat c++ code 3 years ago
layer_norm.cpp feat(mgb/opr): add layernorm forward and backward kernel 3 years ago
linspace.cpp style(all): reformat c++ code 3 years ago
local.cpp style(all): reformat c++ code 3 years ago
local_share.cpp style(all): reformat c++ code 3 years ago
lrn.cpp style(all): reformat c++ code 3 years ago
lsq.cpp style(all): reformat c++ code 3 years ago
mask_conv.cpp chore(mge): update copyright years 4 years ago
matrix_inverse.cpp chore(mge): update copyright years 4 years ago
matrix_mul.cpp style(all): reformat c++ code 3 years ago
mesh_indexing.cpp style(all): reformat c++ code 3 years ago
padding.cpp style(all): reformat c++ code 3 years ago
param_pack.cpp style(all): reformat c++ code 3 years ago
pooling.cpp feat(dnn): use ref ptr interface for all backends 3 years ago
powc.cpp chore(mge): update copyright years 4 years ago
reduce.cpp fix(dnn): correct reduce/argmxx/fakequant calculation with nan 3 years ago
relayout.cpp style(all): reformat c++ code 3 years ago
relayout_format.cpp style(all): reformat c++ code 3 years ago
remap.cpp feat(dnn): add float16 for remap backward 3 years ago
repeat.cpp style(all): reformat c++ code 3 years ago
resize.cpp style(all): reformat c++ code 3 years ago
rng.cpp feat(mgb/opr): add dropout kernel 3 years ago
roi_align.cpp style(all): reformat c++ code 3 years ago
roi_copy.cpp style(all): reformat c++ code 3 years ago
roi_pooling.cpp style(all): reformat c++ code 3 years ago
rotate.cpp style(all): reformat c++ code 3 years ago
sleep.cpp style(all): reformat c++ code 3 years ago
sliding_window_transpose.cpp style(all): reformat c++ code 3 years ago
softmax.cpp feat(mge): add softmax op use cudnn api 3 years ago
split.cpp style(all): reformat c++ code 3 years ago
svd.cpp chore(mge): update copyright years 4 years ago
tensor_remap.cpp style(all): reformat c++ code 3 years ago
tile.cpp style(all): reformat c++ code 3 years ago
timer.h chore(mge): update copyright years 4 years ago
topk.cpp style(all): reformat c++ code 3 years ago
tqt.cpp style(all): reformat c++ code 3 years ago
transpose.cpp style(all): reformat c++ code 3 years ago
type_cvt.cpp feat(dnn/arm): support layout like NHWC channel like broadcast on arm 3 years ago
utils.cpp fix(dnn/cuda): fix int4 epilogue stg bug 4 years ago
utils.h style(all): reformat c++ code 3 years ago
warp_affine.cpp style(all): reformat c++ code 3 years ago
warp_perspective.cpp style(all): reformat c++ code 3 years ago
Delete task

Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.

cancel
confirm

MegEngine 安装包中集成了使用 GPU 运行代码所需的 CUDA 环境，不用区分 CPU 和 GPU 版。 如果想要运行 GPU 程序，请确保机器本身配有 GPU 硬件设备并安装好驱动。 如果你想体验在云端 GPU 算力平台进行深度学习开发的感觉，欢迎访问 MegStudio 平台

C++ Cuda Python C SVG other