ylchan87
ylchan87
To add one point, from gym's [code](https://github.com/openai/gym/blob/4ede9280f9c477f1ca09929d10cdc1e1ba1129f1/gym/spaces/multi_binary.py) for MultiBinary, `spaces.MultiBinary([5, 4, 5])` would be 5x4x5, not 5+4+5 binary variables. In my case the observation is kind of a 10x10 2D...
Despite the doc, turns out gym itself also dislike N-Dim MultiBinary, ``` def flatdim(space): .... elif isinstance(space, MultiBinary): > return int(space.n) E TypeError: int() argument must be a string, a...
Same issue here and Haoshu's fix worked for me
+1 same issue here with Ubuntu 20.04 and gtk-doc-tools 1.32-4
Thanks, tried `stream-sync=0 ` i.e. (restart-ts) but did not help. Also I notice when switched to shmsrc shmsink, the same problem still persist. Not sure if gst-interpipe uses shm under...
+1 for this feature request In my use case, each function correspond to an action that the agent could perform. As the agent interact with the world, - new actions...