SPACY with CUDA-110

[pratikchhapolika]

How to reproduce the behaviour

Your Environment

• Operating System: Jupyter Notebook Server ( Linux)
• Python Version Used:Python 3.6.12 |Anaconda, Inc.| [GCC 7.3.0] on linux
• spaCy Version Used: 3.1.1
• Environment Information: Jupyter( Pytorch with GPU)

+-----------------------------------------------------------------------------+
| NVIDIA-SMI ****   Driver Version: XXXXXX    **CUDA Version: 11.0**     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  Tesla V100-SXM2...  Off  | 00000000:B2:00.0 Off |                    0 |
| N/A   35C    P0    43W / 300W |      0MiB / 32510MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

GPU not recognised by Spacy

!pip install -U pip setuptools wheel --user
!pip install -U spacy[cuda110] --user

import spacy

spacy.prefer_gpu()

False

[adrianeboyd]

Try something like this to hopefully see a more useful error message:

import cupy
arr = cupy.zeros((5,))