erezalg

Hi @IlyaMescheryakov1402 , Good news! ClearML V1.5.0 is out! (pip install clearml==1.5.0) with support to specifying HP sections :D You can do so by calling: @PipelineDecorator.pipeline(args_map={'SectionA':['paramA'], 'SectionB':['paramB]}) This will map...

@davidenitti , We have an RC ready for this issue, install it by running pip install clearml==1.3.2rc1 Let us know if this solves the issue

Hi @davidenitti, Did you get a chance to test whether this issue is resolved or not?

Hi @federicovallucci and @milongo, You guys are right :) This is on our roadmap. We'll update here once it's out!

Hi @derEitel, We're taking a look, we'll update here once we have a fix

@noamzilo, Thanks for the suggestions! You can find a CIFAR example using clearml-data [here](https://github.com/allegroai/clearml/tree/master/examples/datasets) As for improving MNIST, we'd more than appreciate a PR improving it :smiley:!

Hi @Berttwm, Can you try to modify the "max_image_history" in Logger.report_image() and see if it solves it for you? If it does, you can increase the max # of images...

Hi @Berttwm, I tried this simple code ``` from clearml import Task, Logger t = Task.init(project_name='test',task_name='test_report_image') l = t.get_logger() for i in range(0,200): l.report_image(title='test',series='123',iteration=i,local_path='PATH.png'.format(i),max_image_history=10) ``` For me when I change...

@rui1998 Git information is automatically stored once a file, in a git repository calls Task.init(). Is there any specific information you're referring to?

Hi @rui1998, There isn't really a configuration you need to make for it to work. Are you trying to do something specific and are failing to do it? Here is...