AttributeError: module 'cupy' has no attribute 'cublas'

[jayshow03]

Hi there, does anyone know the answer to this question? I tried different versions of cupy-cuda110 but it did not work so far. Below you can see the unexpected information. Hope someone can address it, thanks!
(pytorch) C:\Users\Jayshow\PycharmProjects\spacy_transform>pip list
WARNING: Ignoring invalid distribution -upy-cuda110 (c:\users\jayshow\anaconda3\envs\pytorch\lib\site-packages)
Package Version

---

aiofiles 0.8.0
aiohttp 3.8.1
aiosignal 1.2.0
altair 4.2.0
argon2-cffi 20.1.0
async-generator 1.10
async-timeout 4.0.1
attrs 21.4.0
backcall 0.2.0
bleach 4.1.0
blis 0.7.5
boto3 1.20.24
botocore 1.23.24
brotlipy 0.7.0
cachetools 5.0.0
catalogue 2.0.6
certifi 2021.10.8
cffi 1.15.0
charset-normalizer 2.0.11
click 7.1.2
colorama 0.4.4
cryptography 36.0.0
cupy-cuda110 8.0.0
cymem 2.0.6
debugpy 1.5.1
decorator 5.1.1
defusedxml 0.7.1
Deprecated 1.2.13
docker 5.0.3
entrypoints 0.3
et-xmlfile 1.1.0
fastapi 0.68.2
fastrlock 0.5
filelock 3.4.2
frozenlist 1.2.0
gobbli 0.2.4
google-api-core 1.25.1
google-auth 1.33.0
google-cloud-core 1.7.1
google-cloud-storage 1.41.0
google-crc32c 1.1.2
google-resumable-media 1.3.1
googleapis-common-protos 1.53.0
grpcio 1.42.0
h11 0.13.0
huggingface-hub 0.0.8
humanize 3.14.0
idna 3.3
importlib-metadata 4.8.2
ipykernel 6.4.1
ipython 7.31.1
ipython-genutils 0.2.0
ipywidgets 7.6.5
jedi 0.18.1
Jinja2 3.0.3
jmespath 0.10.0
joblib 1.1.0
jsonschema 3.2.0
jupyter 1.0.0
jupyter-client 7.1.2
jupyter-console 6.4.0
jupyter-core 4.9.1
jupyterlab-pygments 0.1.2
jupyterlab-widgets 1.0.0
langcodes 3.3.0
MarkupSafe 2.0.1
matplotlib-inline 0.1.2
mistune 0.8.4
mkl-fft 1.3.1
mkl-random 1.2.2
mkl-service 2.4.0
msgpack 1.0.3
multidict 5.1.0
murmurhash 1.0.6
nbclient 0.5.3
nbconvert 6.1.0
nbformat 5.1.3
nest-asyncio 1.5.1
numpy 1.22.2
olefile 0.46
openpyxl 3.0.9
packaging 21.3
pandas 1.4.0
pandocfilters 1.5.0
parso 0.8.3
pathy 0.6.1
peewee 3.14.8
pickleshare 0.7.5
Pillow 8.2.0
pip 21.2.2
plac 1.1.3
preshed 3.0.6
prodigy 1.11.6
prometheus-client 0.13.1
prompt-toolkit 3.0.20
protobuf 3.19.1
pyasn1 0.4.8
pyasn1-modules 0.2.8
pycparser 2.21
pydantic 1.8.2
Pygments 2.11.2
PyJWT 2.3.0
pyOpenSSL 22.0.0
pyparsing 3.0.7
pyrsistent 0.18.0
PySocks 1.7.1
python-dateutil 2.8.2
pytz 2021.3
pywin32 227
pywinpty 2.0.2
PyYAML 6.0
pyzmq 22.3.0
qtconsole 5.2.2
QtPy 1.11.2
ray 1.10.0
redis 4.1.3
regex 2021.8.3
requests 2.27.1
rsa 4.7.2
s3transfer 0.5.0
sacremoses 0.0.43
scikit-learn 1.0.2
scipy 1.8.0
Send2Trash 1.8.0
setuptools 58.0.4
shellingham 1.3.1
sip 4.19.13
six 1.16.0
smart-open 5.2.1
spacy 3.2.0
spacy-alignments 0.8.4
spacy-legacy 3.0.8
spacy-loggers 1.0.1
spacy-transformers 1.1.4
srsly 2.4.2
starlette 0.14.2
terminado 0.13.1
testpath 0.5.0
thinc 8.0.13
threadpoolctl 3.1.0
tokenizers 0.10.3
toolz 0.11.2
torch 1.7.1
torchvision 0.8.2
tornado 6.1
tqdm 4.62.3
traitlets 5.1.1
transformers 4.6.0
typer 0.3.2
typing-extensions 3.10.0.2
urllib3 1.26.8
uvicorn 0.13.4
wasabi 0.9.0
wcwidth 0.2.5
webencodings 0.5.1
websocket-client 1.2.3
wheel 0.37.1
widgetsnbextension 3.5.1
win-inet-pton 1.1.0
wincertstore 0.2
wrapt 1.13.3
xlrd 2.0.1
yarl 1.5.1
zipp 3.7.0

(pytorch) C:\Users\Jayshow\PycharmProjects\spacy_transform>python -m spacy train textcatcorpus\config.cfg --output ./textcatoutput --paths.train
textcatcorpus\train.spacy --paths.dev textcatcorpus\dev.spacy --gpu-id 0
C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\numpy_distributor_init.py:30: UserWarning: loaded more than 1 DLL from .libs:
C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\numpy.libs\libopenblas.EL2C6PLE4ZYW3ECEVIV3OXXGRN2NRFM2.gfortran-win_amd64.dll
C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\numpy.libs\libopenblas.GK7GX5KEQ4F6UYO3P26ULGBQYHGQO7J4.gfortran-win_amd64.dll
warnings.warn("loaded more than 1 DLL from .libs:"
ℹ Saving to output directory: textcatoutput
ℹ Using GPU: 0

=========================== Initializing pipeline ===========================
[2022-02-12 20:50:40,286] [INFO] Set up nlp object from config
[2022-02-12 20:50:40,298] [INFO] Pipeline: ['textcat_multilabel']
[2022-02-12 20:50:40,304] [INFO] Created vocabulary
[2022-02-12 20:50:40,306] [INFO] Finished initializing nlp object
[2022-02-12 20:50:41,868] [INFO] Initialized pipeline components: ['textcat_multilabel']
✔ Initialized pipeline

============================= Training pipeline =============================
ℹ Pipeline: ['textcat_multilabel']
ℹ Initial learn rate: 0.0001
E # LOSS TEXTC... CATS_SCORE SCORE

---

⚠ Aborting and saving the final best model. Encountered exception:
AttributeError("module 'cupy' has no attribute 'cublas'")
Traceback (most recent call last):
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\runpy.py", line 194, in _run_module_as_main
return _run_code(code, main_globals, None,
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\runpy.py", line 87, in run_code
exec(code, run_globals)
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\spacy_main.py", line 4, in
setup_cli()
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\spacy\cli_util.py", line 71, in setup_cli
command(prog_name=COMMAND)
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\click\core.py", line 829, in call
return self.main(*args, **kwargs)
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\click\core.py", line 782, in main
rv = self.invoke(ctx)
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\click\core.py", line 1259, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\click\core.py", line 1066, in invoke
return ctx.invoke(self.callback, **ctx.params)
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\click\core.py", line 610, in invoke
return callback(*args, **kwargs)
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\typer\main.py", line 497, in wrapper
return callback(**use_params) # type: ignore
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\spacy\cli\train.py", line 45, in train_cli
train(config_path, output_path, use_gpu=use_gpu, overrides=overrides)
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\spacy\cli\train.py", line 75, in train
train_nlp(nlp, output_path, use_gpu=use_gpu, stdout=sys.stdout, stderr=sys.stderr)
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\spacy\training\loop.py", line 122, in train
raise e
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\spacy\training\loop.py", line 105, in train
for batch, info, is_best_checkpoint in training_step_iterator:
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\spacy\training\loop.py", line 219, in train_while_improving
proc.finish_update(optimizer) # type: ignore[attr-defined]
File "spacy\pipeline\trainable_pipe.pyx", line 252, in spacy.pipeline.trainable_pipe.TrainablePipe.finish_update
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\thinc\model.py", line 327, in finish_update
param, grad = optimizer(
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\thinc\optimizers.py", line 230, in call
gradient = ops.clip_gradient(gradient, self.grad_clip)
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\thinc\backends\cupy_ops.py", line 107, in clip_gradient
grad_norm = cupy.maximum(frobenius_norm(gradient), 1e-12)
File "C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\thinc\backends\cupy_ops.py", line 105, in frobenius_norm
return cupy.cublas.nrm2(X_vec)
AttributeError: module 'cupy' has no attribute 'cublas'

[polm]

Based on the errors and warnings here it seems like there's something wrong with your environment.

WARNING: Ignoring invalid distribution -upy-cuda110 (c:\users\jayshow\anaconda3\envs\pytorch\lib\site-packages)

You should figure out where that's coming from and fix it.

C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\numpy_distributor_init.py:30: UserWarning: loaded more than 1 DLL from .libs:
C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\numpy.libs\libopenblas.EL2C6PLE4ZYW3ECEVIV3OXXGRN2NRFM2.gfortran-win_amd64.dll
C:\Users\Jayshow\anaconda3\envs\pytorch\lib\site-packages\numpy.libs\libopenblas.GK7GX5KEQ4F6UYO3P26ULGBQYHGQO7J4.gfortran-win_amd64.dll
warnings.warn("loaded more than 1 DLL from .libs:"

This makes it look like something is wrong with your numpy or pytorch installs?

Can you start over in a fresh environment and follow the instructions here? You shouldn't have to try different versions of cuda-cupy - you should just get one that matches your cuda version and pip should install the latest compatible one.

[dariusbakunas]

This helped me: explosion/thinc#920

pip install thinc==8.2.3

[Sturmharder]

Thanks!!! That helped me