Martin Marenz

Hi, I want to start contributing to cudf and would like to work on this issue. Should I create one PR per benchmark or put all in one PR?

Hi, I added a few helper function which I would use also for the other groupby benchmarks. PS: Should I add the other `groupby` benchmarks to this PR to reduce...

> Would other benchmarks benefit from those helper functions as well? If not, we can move them to `benchmarks I think so. They could be used in a few of...

Hi, the PR is not ready, I had not enough time the last two days. So no need to review it :smiley:.

Hi @PointKernel , @davidwendt, in the current state of the PR every NVBENCH benchmark has `bytes_per_second` added. However, I'm a bit unhappy with its current state. Specifically, I'd like to...

The proposed changes could potentially result in unpredictable outcomes for torchrl users. Users might employ other TensorDict keys, expecting them to transition from ("next", "some_key") to "some_key" after one step....

You are right. I got the meaning of the keyword wrong. I was also wrong with the BC, the input tensordict will be cloned in all loss_module.forward() instances. So I...