Florian Huber
Florian Huber
I also did not get it to work with conda, even though trying multiple fresh conda environments with different python versions. The Errors I get are also very inconsitent =...
Sorry, I noticed (too late) that the readme mentions running things under WSL under Windows. Using the linux subsystem, the conda install works fine.
Hi @sverhoeven ! I here tried a few things to fix the failing conda verify tests in #230, but with no success. Do you have any further ideas how to...
Outdated --> will be closed
I spend a bit of time to check if I could implement this using `numba` but that didn't work out because of too many non-numba style functions being invovled. Could...
Sound great! Let me know if I can help with anything. Just a minor comment regarding expectations: we optimized the indivual pair-wise score computation quite a bit. But the double-for-loop...
I ran a few more experiments using `joblib`, `dask`, `ray` and `mpire`. So far I only really got the first two to work. Short summary: - Individual score computations seem...
Hi @chetankm1992, thanks for the question. This is something that matchms currently cannot handle properly. The reason is that you simply want to store too many scores at once. Since...
Out of curiosity: What is your aim with this calculation? Are you -in the end- interested in all scores or only the higher ones? If so, keeping only values above...
In principle, using a generator sounds like a good idea. But I foresee that this will be very tricky to implement because of the way the calculations are done. If...