Change attribute name "f1_score" to "f1 score" due to RubixML change

[DerDreschner]

In version 1.0.0-beta1 of RubixML, the attributes in the Report property changed their style from underscore to spaces (see changelog as well as associated commit). This means that the f1_score were always 0.00 since the bump of RubixML dependency in this commit (dev-chore/bump-flysystem-v2.1.1 is based on RubixML v0.4.2). Can't say if that made the training totally broken or not (@ChristophWurst ?).

Example output of a training epoch without change:

Epoch 0: epochs= 330 layers= 2 shuffledRate=0.005 randomRate=2.000, learningRate=0.0070
  Step width for next config neighbor: 0.8
  Training result: f1=0.000000, p(n)=0.972973, r(n)=0.947368, f1(n)=0.000000, p(y)=0.948718, r(y)=0.973684, f1(y)=0.000000, PSR=117/1/234
  Training result: f1=0.000000, p(n)=0.969697, r(n)=0.842105, f1(n)=0.000000, p(y)=0.860465, r(y)=0.973684, f1(y)=0.000000, PSR=117/1/234
  Training result: f1=0.000000, p(n)=0.972973, r(n)=0.947368, f1(n)=0.000000, p(y)=0.948718, r(y)=0.973684, f1(y)=0.000000, PSR=117/1/234
  Training result: f1=0.000000, p(n)=0.970588, r(n)=0.868421, f1(n)=0.000000, p(y)=0.880952, r(y)=0.973684, f1(y)=0.000000, PSR=117/1/234
  Training result: f1=0.000000, p(n)=0.972973, r(n)=0.947368, f1(n)=0.000000, p(y)=0.948718, r(y)=0.973684, f1(y)=0.000000, PSR=117/1/234
  Training result: f1=0.000000, p(n)=0.612903, r(n)=1.000000, f1(n)=0.000000, p(y)=1.000000, r(y)=0.368421, f1(y)=0.000000, PSR=117/1/234
  Training result: f1=0.000000, p(n)=0.972973, r(n)=0.947368, f1(n)=0.000000, p(y)=0.948718, r(y)=0.973684, f1(y)=0.000000, PSR=117/1/234
  Training result: f1=0.000000, p(n)=0.791667, r(n)=1.000000, f1(n)=0.000000, p(y)=1.000000, r(y)=0.736842, f1(y)=0.000000, PSR=117/1/234
  Base cost is 0. Trying to optimize this now …

Example output of a training epoch with change from this PR:

Epoch 0: epochs= 330 layers= 2 shuffledRate=0.005 randomRate=2.000, learningRate=0.0070
  Step width for next config neighbor: 0.8
  Training result: f1=0.947222, p(n)=0.904762, r(n)=1.000000, f1(n)=0.950000, p(y)=1.000000, r(y)=0.894737, f1(y)=0.944444, PSR=117/1/234
  Training result: f1=0.839323, p(n)=0.770833, r(n)=0.973684, f1(n)=0.860465, p(y)=0.964286, r(y)=0.710526, f1(y)=0.818182, PSR=117/1/234
  Training result: f1=0.986840, p(n)=0.974359, r(n)=1.000000, f1(n)=0.987013, p(y)=1.000000, r(y)=0.973684, f1(y)=0.986667, PSR=117/1/234
  Training result: f1=0.960519, p(n)=0.972973, r(n)=0.947368, f1(n)=0.960000, p(y)=0.948718, r(y)=0.973684, f1(y)=0.961039, PSR=117/1/234
  Training result: f1=0.788889, p(n)=0.761905, r(n)=0.842105, f1(n)=0.800000, p(y)=0.823529, r(y)=0.736842, f1(y)=0.777778, PSR=117/1/234
  Training result: f1=0.960519, p(n)=0.972973, r(n)=0.947368, f1(n)=0.960000, p(y)=0.948718, r(y)=0.973684, f1(y)=0.961039, PSR=117/1/234
  Training result: f1=0.960519, p(n)=0.972973, r(n)=0.947368, f1(n)=0.960000, p(y)=0.948718, r(y)=0.973684, f1(y)=0.961039, PSR=117/1/234
  Training result: f1=0.947332, p(n)=0.972222, r(n)=0.921053, f1(n)=0.945946, p(y)=0.925000, r(y)=0.973684, f1(y)=0.948718, PSR=117/1/234
  Base cost is 0.92591180270686. Trying to optimize this now …

Closes #1021.

[ChristophWurst]

Can't say if that made the training totally broken or not (@ChristophWurst ?)

Not in general. Regular model training would still work, just log incorrect statistics. It's the optimization command that is effected, but it's also not executed automatically but only when admins are interested to try it.

[ChristophWurst]

Thank you!!

[ChristophWurst]

/backport to stable31

[github-actions]

Hello there,
Thank you so much for taking the time and effort to create a pull request to our Nextcloud project.

We hope that the review process is going smooth and is helpful for you. We want to ensure your pull request is reviewed to your satisfaction. If you have a moment, our community management team would very much appreciate your feedback on your experience with this PR review process.

Your feedback is valuable to us as we continuously strive to improve our community developer experience. Please take a moment to complete our short survey by clicking on the following link: https://cloud.nextcloud.com/apps/forms/s/i9Ago4EQRZ7TWxjfmeEpPkf6

Thank you for contributing to Nextcloud and we hope to hear from you soon!

(If you believe you should not receive this message, you can add yourself to the blocklist.)