typewriter
typewriter copied to clipboard
logico
An iA Writer inspired color scheme. It has support for vim, vim airline, vim lightline, tmux and urxvt/xterm.
I wonder if there's support for Carla version 9.2 instead of Carla version 8.2 ?
Hi, I'm trying to implement a version of [RND](https://blog.openai.com/reinforcement-learning-with-prediction-based-rewards/#RNDjump) for implicit rewards. I'd like it to be both environment agnostic and agent agnostic. The idea was to implement it as...
A comment from the code: ``` # TODO-remove - this is a temporary flow, used by the trainer worker, duplicated from observe() - need to create # an external trainer...
One notable exception from the Observation spaces as they are is that of arbitrary length sequences, such as nlp or coding problems. The old games that were text IO are...
The `json` files stored in the `experiments` directory are not valid `json` objects. Is there anything preventing us from using valid json? Is there a method available for loading this...
According to the original DQN paper, observation preprocessing should be done as - ` The raw frames are preprocessed by first converting their RGB representation to gray-scale and down-sampling it...
This comment in code indicates that they are supported: https://github.com/NervanaSystems/coach/blob/master/rl_coach/level_manager.py#L50 But don't see any documentation or examples.
Right now Coach saves checkpoints every X seconds. It would be great if I can save checkpoints every X iterations or save checkpoints if it reaches an evaluation reward threshold....
I wonder if there's support for using Imitation Learning API in Carla's environment. For Example, if I try imitation learning on an Atari Environment I use: coach -et rl_coach.environments.gym_environment:Atari -lvl...
Currently Coach supports multiple actors in the multi-node setting. Add support for multiple trainers as well.
