【Help】环境安装报错：安装flash-attn报错，它无法支持910B。当前能支持昇腾910B吗，请问还有其他方法吗？

[Kaimar666]

报错信息：
flash_attn-2.7.4.post1.tar.gz (6.0 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.0/6.0 MB 66.7 MB/s eta 0:00:00
Preparing metadata (setup.py) ... error
error: subprocess-exited-with-error

× python setup.py egg_info did not run successfully.
│ exit code: 1
╰─> [6 lines of output]
Traceback (most recent call last):
File "", line 2, in
File "", line 34, in
File "/tmp/pip-install-_oh5xiov/flash-attn_f996f45be23845c5b7293c3958f895c0/setup.py", line 22, in
import torch
ModuleNotFoundError: No module named 'torch'
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

note: This is an issue with the package mentioned above, not pip.
hint: See above for details.

---

Package Version

---

accelerate 1.0.1
aiofiles 23.2.1
aiohappyeyeballs 2.4.3
aiohttp 3.10.10
aiosignal 1.3.1
annotated-types 0.7.0
anyio 4.6.2.post1
arrow 1.2.3
ascendebug 0.1.0
asttokens 2.4.1
async-timeout 4.0.3
attrs 24.2.0
auto-tune 0.1.0
av 13.1.0
binaryornot 0.4.4
certifi 2024.8.30
cffi 1.15.1.post20240308173724
chardet 5.2.0
charset-normalizer 3.4.0
click 8.1.7
comm 0.2.2
configparser 5.2.0
contourpy 1.3.0
cookiecutter 2.6.0
cryptography 3.4.7
cycler 0.12.1
dataflow 0.0.1
datasets 3.1.0
debugpy 1.8.7
decorator 5.1.1
deepspeed 0.15.4
dill 0.3.8
docstring_parser 0.16
einops 0.8.0
ephemeral-port-reserve 1.1.4
esdk-obs-python 3.21.4
exceptiongroup 1.2.2
executing 2.1.0
fastapi 0.115.4
ffmpy 0.4.0
filelock 3.16.1
fire 0.7.0
fonttools 4.54.1
frozenlist 1.5.0
fsspec 2024.9.0
gradio 4.44.1
gradio_client 1.3.0
h11 0.14.0
hccl 0.1.0
hccl-parser 0.1
hjson 3.1.0
httpcore 1.0.6
httpx 0.27.2
huaweicloudsdkcore 3.1.8
huaweicloudsdkcsms 3.1.8
huggingface-hub 0.26.2
idna 3.10
importlib-metadata 4.4.0
importlib_resources 6.4.5
ipykernel 6.29.5
ipython 8.29.0
jedi 0.19.1
Jinja2 3.1.4
jupyter_client 8.6.3
jupyter_core 5.7.2
kiwisolver 1.4.7
llamafactory 0.9.1
llm-datadist 0.0.1
llm-engine 0.0.1
lxml 5.1.0
ma-cli 1.2.3
markdown-it-py 3.0.0
MarkupSafe 2.1.5
matplotlib 3.9.2
matplotlib-inline 0.1.7
mdurl 0.1.2
mock 4.0.3
modelarts 1.4.20
mpmath 1.3.0
msadvisor 1.0.0
msgpack 1.1.0
multidict 6.1.0
multiprocess 0.70.16
nest-asyncio 1.6.0
networkx 3.4.2
ninja 1.11.1.3
npu-bridge 1.15.0
npu-device 0.1
numpy 1.26.4
op-compile-tool 0.1.0
op-gen 0.1
op-test-frame 0.1
opc-tool 0.1.0
orjson 3.10.11
packaging 24.1
pandas 2.2.3
parso 0.8.4
peft 0.12.0
pexpect 4.9.0
pillow 10.4.0
pip 24.2
platformdirs 4.3.6
prettytable 3.5.0
prompt_toolkit 3.0.48
propcache 0.2.0
protobuf 5.28.3
psutil 6.1.0
ptyprocess 0.7.0
pure_eval 0.2.3
py-cpuinfo 9.0.0
pyarrow 18.0.0
pycparser 2.21
pydantic 2.9.2
pydantic_core 2.23.4
pydub 0.25.1
Pygments 2.18.0
pyparsing 3.2.0
python-dateutil 2.9.0.post0
python-multipart 0.0.17
python-slugify 8.0.4
pytz 2024.2
PyYAML 6.0.2
pyzmq 26.2.0
regex 2024.9.11
requests 2.32.3
requests-futures 1.0.0
requests-toolbelt 0.10.1
rich 13.9.4
ring-flash-attn 0.1.4
ruff 0.7.2
safetensors 0.4.5
schedule-search 0.0.1
scipy 1.14.1
semantic-version 2.10.0
sentencepiece 0.2.0
setuptools 75.1.0
shellingham 1.5.4
shtab 1.7.1
simplejson 3.17.0
six 1.16.0
sniffio 1.3.1
sse-starlette 2.1.3
stack-data 0.6.3
starlette 0.41.2
sympy 1.13.3
tabulate 0.9.0
te 0.4.0
tenacity 8.1.0
termcolor 2.5.0
text-unidecode 1.3
tiktoken 0.8.0
tokenizers 0.19.1
tomlkit 0.12.0
torch 2.1.0
torch-npu 2.1.0.post3
tornado 6.4.1
tqdm 4.67.0
traitlets 5.14.3
transformers 4.43.2
trl 0.9.6
typer 0.12.5
typing_extensions 4.12.2
tyro 0.8.14
tzdata 2024.2
urllib3 2.2.3
uvicorn 0.32.0
wcwidth 0.2.13
websockets 12.0
wheel 0.44.0
xxhash 3.5.0
yarl 1.17.1
zipp 3.7.0

[Eisenhower]

同问

[HaoshengZou]

目前应该是因为昇腾910B不原生支持flash attention。我们目前的实现是用flash attention实现序列并行的。
@Eisenhower @Kaimar666 我们对昇腾910B不太熟悉，你们知道它支持deepspeed吗？如果支持deepspeed，就支持deepspeed Ulysses，我们一直有计划增加通过deepspeed Ulysses实现序列并行。