Franco Peschiera

Yes, I actually cloned the branch and ran the source code when I tested it to be sure it's not using numpy. Curiously, the pulp version I have took less...

The complete code I ran was: ``` import pulp as pl import pysmps.smps_loader as mps file = 'dance_pairing_-pulp.mps.txt' # this runs very fast pul_data = pl.mpslp.readMPS(file, pl.LpMinimize) print(pul_data) # this...

It's entirely possible that I have an old version of your library. It's also possible that I made some (many) mistakes while porting it. All the more reasons to use...

Haha I think maybe we're taking this initial issue too far. Thanks for all the work. I haven't had the time to test it. On the other hand I'm not...

Hello Julian, There is definitely a need for an mps reader, yes. I haven't yet tested the new mps_reader.py file. As soon as I do that I'll let you know....

Hello again. Thanks both @jmaerte @MLopez-Ibanez . Indeed, as @MLopez-Ibanez said, my plate is quite full (and then I added holidays). I've pulled the new version of the branch and...

Sorry, I just realized I was using the incorrect import. If I change my second line to `import pysmps.mps_loader as mps` then the speed is the same. I will continue...

Hello again, I've adapted the code in the following pulp branch: https://github.com/coin-or/pulp/tree/pysmps Tests still fail to run. If you want to try the tests you can clone the repository and...

well, PULP_CBC_CMD calls the "embedded" binary, while the COIN_CMD API calls whatever installed binary you have in your PATH. In your case, I imagine the tests that fail are the...

please run the black linter as per instructions: https://coin-or.github.io/pulp/develop/contribute.html#applying-the-black-linter-formatter (and check any other step you may be missing). I'll try to review the changes soon.