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.
 
 
 
 
 
 
chenjiahui d17cd60d3b feat(imperative): add pixel_shuffle opr 3 years ago
..
qat docs(docstring): transfer to google style 3 years ago
quantized docs(docstring): transfer to google style 3 years ago
__init__.py feat(mge): add imperative pad 3 years ago
activation.py docs(docstring): transfer to google style 3 years ago
adaptive_pooling.py docs(docstring): transfer to google style 3 years ago
batch_matmul_activation.py docs(docstring): transfer to google style 3 years ago
batchnorm.py docs(docstring): transfer to google style 3 years ago
concat.py docs(docstring): transfer to google style 3 years ago
conv.py docs(docstring): transfer to google style 3 years ago
conv_bn.py docs(docstring): transfer to google style 3 years ago
deformable_psroi_pooling.py feat(mge/quantization): add name for quantized module 4 years ago
dropout.py docs(docstring): transfer to google style 3 years ago
elemwise.py docs(docstring): transfer to google style 3 years ago
embedding.py docs(docstring): transfer to google style 3 years ago
external.py feat(imperative/opr): add extern c opr and support dump 3 years ago
identity.py refactor(mge/functional): move functional api 4 years ago
init.py docs(docstring): transfer to google style 3 years ago
linear.py docs(docstring): transfer to google style 3 years ago
module.py feat(module): add tensors and named_tensors 3 years ago
normalization.py feat(mgb): add megbrain layer norm opr with subgraph 3 years ago
padding.py style(mge): change the pad opr doc and format 3 years ago
pixel_shuffle.py feat(imperative): add pixel_shuffle opr 3 years ago
pooling.py docs(docstring): transfer to google style 3 years ago
quant_dequant.py docs(docstring): transfer to google style 3 years ago
sequential.py docs(docstring): transfer to google style 3 years ago
sliding_window.py docs(docstring): transfer to google style 3 years ago

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