@@ -129,15 +129,14 @@ machine learning library, [Thinc](https://thinc.ai). For GPU support, we've been
129
129
grateful to use the work of Chainer' s [CuPy](https://cupy.chainer.org) module,
130
130
which provides a numpy-compatible interface for GPU arrays.
131
131
132
- spaCy can be installed on GPU by specifying `spacy[cuda]`, `spacy[cuda90]`,
133
- `spacy[cuda91]`, `spacy[cuda92]`, `spacy[cuda100]`, `spacy[cuda101]`,
134
- `spacy[cuda102]`, `spacy[cuda110]`, `spacy[cuda111]` or `spacy[cuda112]`. If you
135
- know your cuda version, using the more explicit specifier allows cupy to be
136
- installed via wheel, saving some compilation time. The specifiers should install
132
+ spaCy can be installed for a CUDA-compatible GPU by specifying `spacy[cuda]`,
133
+ `spacy[cuda102]`, `spacy[cuda112]`, `spacy[cuda113]`, etc. If you know your
134
+ CUDA version, using the more explicit specifier allows CuPy to be installed via
135
+ wheel, saving some compilation time. The specifiers should install
137
136
[`cupy`](https://cupy.chainer.org).
138
137
139
138
```bash
140
- $ pip install -U %%SPACY_PKG_NAME[cuda92 ]%%SPACY_PKG_FLAGS
139
+ $ pip install -U %%SPACY_PKG_NAME[cuda113 ]%%SPACY_PKG_FLAGS
141
140
```
142
141
143
142
Once you have a GPU-enabled installation, the best way to activate it is to call
0 commit comments