Maximilian Du
Maximilian Du
Awesome! Thanks.
Does this work? https://drive.google.com/file/d/1QyN5YyZYhzJRAUcTPmRGmebkpyh_9-hy/view?usp=sharing It's from the can pick-place environment. I just ran the `run_trained_agent.py` script using this checkpoint from the repository, and the problem is there
Got it, ok! Yeah, so if I omit this reloading, a policy trained on the same data (which used this reloading during collection) has performed upwards of 10% worse. From...
Let me give this a try with one of the public datsets. I'll keep you posted!
I'll close this for now--I think it was a quirk of our own data, as I wasn't able to reproduce it on the public dataset. If something else comes up,...