Hello.
I have an issue where clearml doesn't read from uncommitted changes. I changed the batch size locally and started training. When I check it in Execution tab, I get the changes.
- batchsize: 22
+ batchsize: 4
22 is committed and 4 is not committed (only local). When I start training, I get batchsize of 22. In order to make the batch size 4, I have to commit it as so and push to GitHub. What is controlling the uncommitted changes use. Do I have to do anything to make it use the code in uncommitted changes? It was working for me before. I have changed the code some what and I get this issue.
Hi @<1523704674534821888:profile|SourLion48> , I'd suggest connecting your batch size as a configuration parameter of the experiment, for example using argparser, and then regardless of the committed or uncommitted code, you will be able to control this value through the configuration section.
What do you think?
You are right. But, I have to start it from draft in the UI to do that, right? I mean, clone and restart.
