-
-
Notifications
You must be signed in to change notification settings - Fork 4.6k
Closed
Labels
feat / serializeFeature: Serialization, saving and loadingFeature: Serialization, saving and loadinggpuUsing spaCy on GPUUsing spaCy on GPU
Description
I am trying to load spacy pickled model using spacy.load("path/model.pkl") as I wanted to use gpu.
spacy.prefer_gpu()
spacy.load("path/model.pkl")
Above code giving me error.
OSError: [E050] Can't find model '<spacy.lang.en.English object at 0x7e8e7d13ef20>'. It doesn't seem to be a Python package or a valid path to a data directory.
The data type of the model is spacy.lang.en.English, despite this I am getting error. Could you please suggest - how to use gpu to load the model to gpu and how to use it efficiently for inference?
Your Environment
- Operating System: Windows
- Python Version Used: 3.9
- spaCy Version Used: spacy_transformers==1.3.3, spacy==3.6.1
- Environment Information:
Metadata
Metadata
Assignees
Labels
feat / serializeFeature: Serialization, saving and loadingFeature: Serialization, saving and loadinggpuUsing spaCy on GPUUsing spaCy on GPU