My understanding is that in Task.init, you have a reuse_last_task_id (or similar name) that defaults to True.. In that case if your experiment wasn't "published" it will be overwritten, (based on project and experiment name). However, if you do publish it, a new experiment would be created
Hi Everyone, I'Ve Seen That When Re-Running A Script It Sometimes Overwrites A Previous Task In The Dashboard Instead Of Creating A New Task. How Does Clearml Decides Whether To Create A New Task Or Overwrite An Existing?
one year ago
10 months ago