ByteDance-Seed / ByteDance-Seed/Bagel
imports doesnt works
- Dominant language
- Python
- Stars
- 6.2k
- Forks
- 545
- PR merge metrics
- No merged PRs in 30d
Description
running on this image:runpod/pytorch:2.2.0-py3.10-cuda12.1.1-devel-ubuntu22.04
on runpod,
after installing requrements.txt i face this error:ImportError: /usr/local/lib/python3.10/dist-packages/flash_attn_2_cuda.cpython-310-x86_64-linux-gnu.so: undefined symbol: _ZN3c104cuda9SetDeviceEi
altho flash_attn didnt give any error during installation. when i try to build flash attn 2.5.8 manually, it runs out of ram(which is very strange becuase ram is 125gb)
Contributor guide
No contributing guide indexed for this repository
Research direction
Start by reproducing the ImportError in the runpod/pytorch:2.2.0-py3.10-cuda12.1.1-devel-ubuntu22.04 image after installing requirements.txt. Compare the installed flash-attn, PyTorch, Python, and CUDA environment, then investigate why building flash-attn 2.5.8 exhausts the reported 125 GB of RAM. Done means flash_attn imports successfully in the stated environment without a manual build failure.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- devops, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100