Evaluation of the trainings
I have a sample and want to check the performance of the taggers for which the values are already stored in the files
If you want to just check the performance of one or more taggers which are already present in the
.h5 files, you can
follow the steps explained here.
I have a new sample and I want to evaluate the performance of my trained tagger on that
To do that, you need to follow these steps:
- Add a new entry to
test_filesin your train config file with the path to the sample and the cuts you want to apply (Wildcards are supported).
- Re-run the
evaluate_model.pyscript (see here for explanation how to do that).
- In your results files is now a new entry with the name of the entry you set in
- You should be able to plot the results now using the
plotting_umami.pyscripts (see here for explanation how to do that) or your personal