Jongwook Choi

Improved python folding -- the whole function/class is now foldable, rather than its body only (which should also be fixed in the upstream soon).

This is great, thanks! But it looks like showing a single-core CPU usage. For example, if there is 8 cores (threads), the value can range between 0-800, which I don't...

@vwxyzjn This is a preview of `representative_fn` for `Hypothesis.plot` (#4). I'll need to add more test cases but the API should be usable at the moment. Feedbacks would be appreciated!

As per the docstring, you can use `representative_fn=lambda h: h.grouped.median()` more simply. When a custom representative value is used, it will represent the radius of the error band centered at...

That's a good idea. I am going to add `representative_fn=...` argument (working now).

Hi Costa, this is really a great idea and thanks for your efforts on maintaining the benchmark results. I'd like to use the cleanRL benchmarks data for the example notebook...

Thanks, this looks like a good option to add. I have two thoughts: First, it would be great if one can list multiple color entries in a single flag (with...

Thanks for reporting this. Would you able to get any relevant information from raw `nvidia-smi` command? If you have any good reference/documentation about it, that'd be also helpful.

The issue is that the `pynvml` library we are relying on is not aware of [MIG](https://docs.nvidia.com/datacenter/tesla/mig-user-guide/index.html). One dirty but quick workaround would be parsing `nvidia-smi` output, but this doesn't seem...

I realized that there is now an *NVIDIA official* python nvml binding **pynvml**: https://pypi.org/project/nvidia-ml-py/ (being actively maintained) which has MIG support. So we'll have to switch to this `pynvml` library,...