-
### bug描述 Describe the Bug
HEAD is now at 7cc9639 Merge pull request #136 from Cyan4973/dev
[540/2110] Performing build step for 'extern_xxhash'
[1/4] Building C object CMakeFiles\xxhsum.dir\D_\CV\…
-
### 🐛 Describe the bug
I found that `torch.linalg.inv` is relatively slow for many (= 100,000) small (< 32 x 32) matrices compared to CuPy.
The complete benchmark code is uploaded here: https://gi…
-
## 🚀 Feature
Implement `torch.Tensor.__array_interface__` attribute as defined by [Array Interface](https://numpy.org/doc/stable/reference/arrays.interface.html)
## Motivation
The array…
pearu updated
3 years ago
-
### 问题描述 Issue Description
按照[官网文档](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/install/compile/arm-compile.html)进行操作，到make（第9步）报错：
root@b6cb716474d7:/opt/environments/Paddle/bui…
-
I have tried to use my code, which works perfectly fine offline, on Google Colab. In an attempt to convert data from CPU to GPU for ML training using cuML i get an Error.
Here is the part of my co…
-
SubmoduleAwareBuilder incorrectly works with "deep" submodules and when len(submodules) > 2
Lets consider https://github.com/microsoft/onnxruntime.
Look, at submodule lists: modules can include …
-
Currently, switching between lazy and eager can be a huge overhead even when using the same device. This is mainly due to the ir graph execution and the conversion of tensor device types. However, the…
-
For semi-supervised dimensionality reduction using UMAP, should I follow the same guidelines described here: [Using Partial Labelling (Semi-Supervised UMAP)](https://umap-learn.readthedocs.io/en/lates…
-
We need a protocol similar to the `__cuda_array_interface__` (CAI) protocol for DPPY. @oleksandr-pavlyk has already drafted an initial version of a CAI-like protocol called `__sycl_usm_array_interface…
-
It's a well-known fact that many convolutions can be thought of as a direct matrix multiplication(Im2Col and more subtle ideas). cuDNN white-paper directly states that NVIDIA developers use precisely …