Skip to content

Fix error when installing in envs that don't already have dependencies#5

Open
prescod wants to merge 1 commit into
Dao-AILab:mainfrom
prescod:main
Open

Fix error when installing in envs that don't already have dependencies#5
prescod wants to merge 1 commit into
Dao-AILab:mainfrom
prescod:main

Conversation

@prescod
Copy link
Copy Markdown

@prescod prescod commented Dec 8, 2023

Fix #2

@prescod prescod changed the title Main Fix error when installing in envs that don't already have dependencies Dec 8, 2023
@lczyk
Copy link
Copy Markdown

lczyk commented Jan 24, 2024

run into the same thing. bump 👍

@alexc-motif
Copy link
Copy Markdown

@tridao can we merge this? cannot install on certain images otherwise

@tridao
Copy link
Copy Markdown
Member

tridao commented Jan 31, 2024

I've experimented w pyproject.toml before and it works for some cases but for other setups it downloads and installs a different version of pytorch, making installation very slow.
Packaging with torch and CUDA is a headache.
I use pip install causal-conv1d --no-build-isolation.

@alexc-motif
Copy link
Copy Markdown

k, thanks for the context

@BenTenmann
Copy link
Copy Markdown

I've experimented w pyproject.toml before and it works for some cases but for other setups it downloads and installs a different version of pytorch, making installation very slow. Packaging with torch and CUDA is a headache. I use pip install causal-conv1d --no-build-isolation.

Hi @tridao, I appreciate this hotfix! But for our use-case, we cannot use --no-build-isolation unfortunately. Any chance we could merge this PR? We would love to use this tool to speed up our inference runs, but just can't because we cannot lock our requirements.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Cannot install on ends that do not have dependencies installed

5 participants