mikemccand commented on PR #14178: URL: https://github.com/apache/lucene/pull/14178#issuecomment-2833392738
> > Do commercial fees apply to conda-forge? > > No. Our commercial fees do not apply to the user-uploaded packages at [anaconda.org](http://anaconda.org/), which includes conda-forge. We do not build these packages. We host them as a free service to the community – something we’ve done for over 10 years. While commercial fees don’t apply, our Terms of Service do still apply to conda-forge, as they do for any third-party content we host, which is typically the case when an internet service provider hosts third-party content > > Based on this, I assume the [`pytorch`](https://anaconda.org/pytorch) channel is free-to-use (just like [`conda-forge`](https://anaconda.org/conda-forge)), and I did not find any specific licenses on https://pytorch.org that limit its use, _but inputs are appreciated -- I'm not familiar with this!_ Right, it looks like Anaconda is hosting "non-default" channels for free (Anaconda commercial license does not apply) for the Conda community. It also looks like the `pytorch` channel is such a non-default channel, and its artifacts are likely managed by the pytorch dev community. But I also cannot find any license governing the artifacts that are pushed into the `pytorch` channel. One big part of the complexity of building these packages is the low-level optimizations based on which "bare metal" you are running on -- CPU vs GPU (with or without Nvidia CUDA/Intel GPU?), if CPU which one (Intel vs AMD are compiled differently)... look at the instructions for [building all of pytorch from source](https://github.com/pytorch/pytorch?tab=readme-ov-file#installation)! Also, at the end of all of this, say I am just a developer wanting to benchmark this cool Faiss Lucene Codec (I am actually)... and I install Faiss 1.11.0 via using `miniforge` via Anaconda's `pytorch` channel, I know I can pick `faiss-cpu` or `faiss-gpu` or simply `faiss`, but how do I know whichever I pick is optimized properly for my particular env? I would run on an `AMD Ryzen Threadripper 3990X` ... was `faiss-cpu` compiled with the optimized deps for that CPU? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
