Rename AsyncSamplesOptimizer -> AsyncReplayOptimizer
Add AsyncSamplesOptimizer that implements the IMPALA architecture
integrate V-trace with a3c policy graph
audit V-trace integration
benchmark compare vs A3C and with V-trace on/off
PongNoFrameskip-v4 on IMPALA scaling from 16 to 128 workers, solving Pong in <10 min. For reference, solving this env takes ~40 minutes for Ape-X and several hours for A3C.
This also removes the async resetting code in VectorEnv. While that improves benchmark performance slightly, it substantially complicates env configuration and probably isn't worth it for most envs.
This makes it easy to efficiently support setups like Joint PPO: https://s3-us-west-2.amazonaws.com/openai-assets/research-covers/retro-contest/gotta_learn_fast_report.pdf
For example, for 188 envs, you could do something like num_envs: 10, num_envs_per_worker: 19.
The dict merge prevents crashes when tune is trying to get resource requests for agents and you override a config subkey. The min iter time prevents iterations from getting too small, incurring high overhead. This is easy to run into on Ape-X since throughput can get very high.
We should use episode ids instead of the timestep to determine when sequences should be cut, since when batches are concatenated, increasing t does not guarantee we are part of the same episode.
* Prevent hasher from running out of memory on large files
* dump out keys
* only print if failed
* remove debugging
* Fix lint error. Reverse adding newline.
* Raise application level exception for actor methods that can't be executed and failed tasks.
* Retry task forwarding for actor tasks.
* Small cleanups
* Move constant to ray_config.
* Create ForwardTaskOrResubmit method.
* Minor
* Clean up queued tasks for dead actors.
* Some cleanups.
* Linting
* Notify task_dependency_manager_ about failed tasks.
* Manage timer lifetime better.
* Use smart pointers to deallocate the timer.
* Fix
* add comment
Using the actual batch size reduces the risk of mis-accounting. Here, we under-counted samples since in truncate_episodes mode we were doubling the batch size by accident in policy_evaluator.
This adds a simple DQN+PPO example for multi-agent. We don't do anything fancy here, just syncing weights between two separate trainers. This potentially is wasting some compute, but is very simple to set up.
It might be nice to share experience collection between the top-level trainers in the future.
* Use absolute path to get to thirdparty dir
In case this script is executed from a different directory than the Ray's directory, the `pushd` will fail. This commit uses absolute path to `thirdparty` directory.
* Update setup_thirdparty.sh
