Comments (8)
Here https://pytorch.org/get-started/locally/ it says:
conda install pytorch torchvision -c pytorch
# MacOS Binaries dont support CUDA, install from source if CUDA is needed
Looking at https://www.google.com/search?q=pytorch+macos+cuda the install from source works. and that pytorch has no intention currently to support binaries.
But what I meant in my original reply is that fastai-v1 has no direct dependency on cuda.
fastai-v1 supports whatever pytorch supports (with windows support catching up that is). FWIW, we run a 100% success test suite on CI for MacOS/conda for cpu. MacOS/pip has incompatibility issues between pytorch and numpy. pytorch/pytorch#14359. We have no CI w/ GPU capacity to test with.
If, however, you're referring to fastai-0.7x, then I don't know. I doubt this will ever change since pytorch-0.3 is not being in active development.
from fastai.
Similar issues for me on MacOS
ResolvePackageNotFound:
- cuda90
from fastai.
Same for me on MacOS...
ResolvePackageNotFound:
- cuda90
from fastai.
#88
This issue is similar to this one.
@jph00 has already closed the previous issue.
from fastai.
I can't help with MacOS sorry. If anyone knows how to get it working on MacOS, a PR would be welcomed.
from fastai.
Ran into a similar issue for MacOS. Cuda90 is not supported on Mac, only on Windows and Linux. You can verify package compatibility by searching https://anaconda.org/
from fastai.
This is a year old thread, and this information is no longer correct.
MacOS is partially supported, i.e. not 100% mainly due to issues in pytorch for which fastai has no control over. e.g, this one: pytorch/pytorch#14359
from fastai.
@stas00 Good to hear. Any idea how to get around cuda90 not being available on MacOS? It's a dependency in the fastai env and it's not supported on MacOS.
from fastai.
Related Issues (20)
- DataLoader's function create_item fails when the dataset is Pandas Dataframe
- setting emb_szs colab_learner is ignored causing function call to exit in error
- get_preds gives me non-deterministic results when using dls.train
- Unable to install fastbook when running Jupyter locally
- Found a possible security concern
- '_FakeLoader' object has no attribute 'pin_memory_device'
- Unable to deploy on Sagemaker
- Cannot scroll on image cleaner HOT 1
- get_image_files providing error message
- get_y should be changed to 'category' from 'label'
- AssertionError when trying to predict image data with FastAI HOT 2
- Training multiple FasiAI models and validating using dataframe dataloader(): HOT 1
- can find the rename_extracted funtion HOT 1
- Unable to process Multi-GPU Training
- Want to ask why I use DataBlock to do NLP task,I have many "xxboxs xxunk" in my data? HOT 1
- Issue with setting the number of workers
- Documentation on which is the actual label vs predicted in learn.show_results() is missing HOT 2
- how to use in google colab?sorry I don't find the entrance,I am a new learner
- Fastai version 2.7.14 produces "AttributeError: device" when running .predict HOT 1
- Notebook_launcher: Importing libraries initializes CUDA
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from fastai.