Hi @<1523701070390366208:profile|CostlyOstrich36> ,
Here is the snippet and the original histogram:

for metric in evaluator.metrics:
print(f"reporting plots for metric {metric}")
task.logger.report_single_value("avg_" + metric, float(results[metric].mean()))
plt.figure()
plt.hist(results[metric].dropna(), bins=25)
fig = plt.gcf()
task.logger.report_matplotlib_figure(metric, metric, fig)

Hi @<1582179661935284224:profile|AbruptJellyfish92> , how do the histograms look when you're not in comparison mode?

Can you provide a self contained snippet that creates such histograms that reproduce this behavior please?

Hi @<1582179661935284224:profile|AbruptJellyfish92> , looks like a bug. Can you please open a GitHub issue to follow up on this?

@<1523701070390366208:profile|CostlyOstrich36> None

That is more relevant for "clearml-web", right?

In addition, when I report a histogram with many bins (for example, 60) i get the same result of thin bars:
Seems like a problem with the configurations of the UI plots in general