Kristian Hartikainen
Kristian Hartikainen
Environment sampling seems to be pretty slow especially on image environments. We should think about ways to better parallelize things to speed it up.
Gym environments return extra info values at each training step. Those infos should be saveable in the replay pools.
The code exits with error when I tried running with humanoid run task because of this line https://github.com/rail-berkeley/softlearning/blob/fb70abc6a43a0ef794c3df7e744500d58bca9416/softlearning/environments/adapters/dm_control_adapter.py#L123 The reason is that the 'head_height' attribute of the observation has 0...
For some reason, the training is currently tens of percents slower (in terms of *wall clock time*) than it was prior to the latest refactor. Need to figure out what...
Hey, I recently had several cases where my `SSH_TTY` environment variable was unset in my emacs even though `tmux show-environment SSH_TTY` was correctly pointing to `/dev/pts/0`. This breaks `clipetty-cut` and...
Adds a more informative samples per second logging for `reverb_dataset.py`. Before: ``` Iteration 0 finished in 0.8698790073394775s Iteration 1 finished in 0.9794516563415527s Iteration 2 finished in 0.9291434288024902s ``` After: ```...
Hey, I'm setting up a custom ACME learning algorithm and currently the learning seems to be quite heavily bottlenecked by the reverb dataset sampling. I tried to understand the reverb...
I noticed some differences between the mjcf xml schema and the [documentation's XML Reference](https://mujoco.readthedocs.io/en/stable/XMLreference.html). This fixes two of these differences: 1) `sensor/user` element's attribute defaults. The corresponding change to mujoco...
The `link20_imp_offpolicy` dimensions seemed to be off. This should fix them.
The final error message ("Error occurred while downloading...") is currently in the `progn`-block, causing the `dl/download-default-config` function to always error. This moves the error outside the `progn`-block and should fix...