Nintorac
Nintorac
oh yea, my bad..for some reason I had in my head `__eq__` was implemented for `Track` looks like `__eq__` is implemented for `BaseMessage` so would it be sufficient to for...
I am using https://pypi.org/project/aws-mfa/ This inserts temporary creds into the credentials file upon authentication and then after they expire s3fs fails to work. I would consider this a fairly typical...
Still getting this issue in MLFlow 1.4. Is the situation improved by using a database backend?
You could use pytorchs fork rng and set seed functions to achieve this though it would be good to have this functionality built in for some use cases ``` import...
To enable remote kernels looks fairly straight forward. I think it should be possible to use a ray remote actor to proxy the necessary ports from the ray client to...
Isolation is more tricky I think, ray has namespaces but as far as I know these still share the filesystem so anything written can be read by other ray jobs....
Does it make sense to have `RayWorker` as it's own custom resource here?
seems to be because `torch.utils.cpp_extension.CUDA_HOME` is returning something along the lines of `/path/to/sym:/path/to/actual` in the `setup.py` In the `setup.py` I added a `torch.utils.cpp_extension.CUDA_HOME.split(':')[0]` which allowed the installation to go further...
running `export CUDA_HOME=/usr/local/cuda` in bash seems to do the trick
> My guess is you're using a profile_name and not environment variables? correct nice, hopefully that lands soon then :)