issue-tracking icon indicating copy to clipboard operation
issue-tracking copied to clipboard

Questions, Help, and Issues for Comet ML

Results 114 issue-tracking issues
Sort by recently updated
recently updated
newest added

[Hydra](https://github.com/facebookresearch/hydra) is a popular framework for composing configuration files. Its typical usage involves using a decorator that points to your configuration yaml file on your main function. However, when this...

bug
sdk

I've got a big dict of hyperparameters and logging it at the end takes a lot of space But I still want to see metrics values So I want to...

feature request

When I click on the refresh button, even if I had some of the experiments hidden, the hidden/visible state is not saved, and all experiments are then shown. This doesn't...

bug

I have the following problem running on ddp mode with cometlogger. When I detach the logger from the trainer (i.e deleting`logger=comet_logger`) the code runs. ``` Exception has occurred: AttributeError Can't...

question

Please remove the annoying warning to update comet ml package. ``` COMET INFO: old comet version (3.1.9) detected. current: 3.1.11 please update your comet lib with command: `pip install --no-cache-dir...

feature request
sdk

The Filter by step slider is really useful, however it's a petty that one can only use it when the figure is not opened / maximised. It would be nice...

feature request

The Duration column in project summary page isn't useful for any resumed experiments. To be specific, the formula for the duration column seems to be: Duration = experiment.EndTime - experiment.StartTime...

feature request

If you call `API.get_metrics_summary` with the name of a particular metric, and the experiment has data for that metric, a dictionary is returned. If the experiment lacks data for that...

feature request
sdk

Using comet-ml 3.1.7 on OSX. The following snippet, with some fit method that returns a score ```python config = { "algorithm": "bayes", "parameters": { "learning_rate": { "type": "float", "min": 0.00,...

bug

I am running my scripts on a SLURM-scheduled cluster where the compute nodes don't have internet access. The training script works just fine: I can use an `OfflineExperiment`. But the...

feature request