Anjum Sayed

I can confirm that the same issue exists using PyTorch Lightning. Steps to reproduce: 1. Open getting_started_object_detection.ipynb 2. Add `precision=16` to the `pl.Trainer` args

@DontGiveUpEasily See my comment here: https://github.com/pemami4911/deep-rl/issues/2#issuecomment-400929047 Ideally, the OU noise needs to be decayed so that the actions don't have noise added after convergence

I think `model.plot();` is fixed in master

Thought I might answer my own question now that I've got something that works. It's possible to concatenate multiple `SpatioTemporalDataset` objects using something like this (I have each experiment as...

Hi @ainoam - no, there is already a default Hydra configuration, I was just trying to override certain parameters. A simplified use case is as follows; I'd like to have...

An update on this - there is an undocumented way of doing this that my colleague discovered: `--args overrides="[dataset.start=2025-01-01, dataset.end=2025-04-01]"` Note the double quotes are required. Might be worth adding...