-
### Description of the bug | 错误描述
完成本地部署后提取文档出现错误
2024-11-19 19:01:46.298 | INFO | magic_pdf.pdf_parse_union_core_v2:pdf_parse_union:647 - page_id: 0, last_page_cost_time: 0.0
2024-11-19 19:01:…
-
I am installing flash-attn in image, container environment as follow:
```
Ubuntu 16.04.6
pytorch image: nvcr.io/nvidia/pytorch: 22.04-py3
PyTorch Version 1.12.0a0+bd13bc6
CUDA 11.6
My card is V1…
-
### Describe the bug
The required subgroup size kernel attribute is incorrectly reported on cuda and hip devices. When checking what the compile subgroup size is of a kernel that had the required s…
-
dont want to DOS them and get blocked
-
### Describe the bug
The test, `read_write_unsampled_semaphore`, non-deterministically fails with no useful output.
```
FAIL: SYCL :: bindless_images/dx12_interop/read_write_unsampled_semaphor…
-
```L 01/24/2024 - 07:57:43: [SM] Unable to load extension "accelerator.ext": /home/sappho/tfTEST/tf2/tf/addons/sourcemod/extensions/accelerator.ext.so: undefined symbol: _ZN15google_breakpad16Exceptio…
-
**Describe the Bug**
The `--threads` option for `nvcc` is hardcoded to `4`, which makes building NVIDIA apex on certain environments impossible, specially from a `docker build` on a CI/CD environment…
-
Emitting ninja build file E:\Model\SparseInst-main\detectron2\build\temp.win-amd64-cpython-38\Release\build.ninja...
Compiling objects...
Allowing ninja to set a default number of workers... (overri…
-
The same problem described in [https://github.com/statsmodels/statsmodels/issues/3783](url) happened again when i was trying to use CompareMeans.summary().
Here's my code:
`import numpy as np
impor…
-
@zdennis reported this issue and I've had the same problem getting it up and running. Bundler seems to install everything fine and the deploy just blows up looking for `ey-enzyme`
Relevant backlog:
…