# Ray 0.8.7

- Product: Ray (https://whatsnew.fyi/product/ray)
- Vendor: Anyscale
- Date: 2020-08-13
- Version: 0.8.7
- Original notes: https://github.com/ray-project/ray/releases/tag/ray-0.8.7
- Permalink: https://whatsnew.fyi/product/ray/releases/0.8.7

What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog.

Reuse: the summaries, labels and curation here are © What's New. Quote freely with attribution and a link back; wholesale republication of the corpus is not permitted — terms: https://whatsnew.fyi/terms. The vendors' own release notes remain their publishers'.

---

- **changed** — ObjectIDs are now called ObjectRefs
- **changed** — Ray Autoscaler is now called Ray Cluster Launcher
- **added** — Ray Cluster Launcher has a new colorful, concise output style available with the --log-new-style flag
- **added** — Windows is now officially supported by RLlib
- **added** — Add ray status debug tool and ray --version command
- **added** — Add option for continuous sync of file mounts
- **added** — Option to automatically reconstruct objects stored in plasma after a failure
- **added** — Support Python type checking for Ray tasks
- **added** — New algorithm: Model-Agnostic Meta-Learning (MAML)
- **added** — New algorithm: Model-Based Meta-Policy-Optimization (MB-MPO)
- **added** — Native TensorFlow 2.x support with framework="tf2" configuration option
- **added** — DQN PyTorch support for full Rainbow setup including distributional DQN
- **added** — Python type hints for Policy, Model, Offline, Evaluation, and Env classes in RLlib
- **added** — Tune Function API now supports checkpointing and is usable with all search and scheduling algorithms
- **added** — DistributedTrainableCreator wrapper for distributed parameter tuning with multi-node DistributedDataParallel models
- **added** — Tune now provides a Scikit-learn compatible wrapper for hyperparameter tuning
- **added** — Integration and tutorial for using Ray Tune with Weights and Biases
- **added** — Serve now starts one HTTP server per Ray node for horizontal scalability
- **added** — serve.shadow_traffic(endpoint, backend, fraction) duplicates and sends a fraction of incoming traffic to a specific backend
- **added** — serve.shutdown() cleanup the current Serve instance in Ray cluster
- **deprecated** — Deprecated Policy Optimizer package in favor of new distributed execution API

Highlight
---------
- Ray is moving towards 1.0! It has had several important naming changes.
    - `ObjectID`s are now called `ObjectRef`s because they are not just IDs. 
    - The Ray Autoscaler is now called the Ray Cluster Launcher. The autoscaler will be a module of the Ray Cluster Launcher.
- The Ray Cluster Launcher now has a much cleaner and concise output style. Try it out with `ray up --log-new-style`. The new output style will be enabled by default (with opt-out) in a later release.
- Windows is now officially supported by RLlib. Multi node support for Windows is still in progress.

Cluster Launcher/CLI (formerly autoscaler)
--------------------------------------------
- **Highlight:** This release contains a new colorful, concise output style for `ray up` and `ray down`, available with the `--log-new-style` flag. It will be enabled by default (with opt-out) in a later release. Full output style coverage for Cluster Launcher commands will also be available in a later release. (#9322, #9943, #9960, #9690)
- Documentation improvements (with guides and new sections) (#9687
- Improved Cluster launcher docker support (#9001, #9105, #8840)
- Ray now has Docker images available on Docker hub. Please check out the [ray image](https://hub.docker.com/u/rayproject/ray) (#9732, #9556, #9458, #9281)
- Azure improvements (#8938)
- Improved on-prem cluster autoscaler (#9663)
- Add option for continuous sync of file mounts (#9544)
- Add `ray status` debug tool and `ray --version` (#9091, #8886).
- `ray memory` now also supports redis_password (#9492)
- Bug fixes for the Kubernetes cluster launcher mode (#9968)
- __Various improvements:__ disabling the cluster config cache (#8117), Python API requires keyword arguments (#9256), removed fingerprint checking for SSH (#9133), Initial support for multiple worker types (#9096), various changes to the internal node provider interface (#9340, #9443)

Core
-----
- Support Python type checking for Ray tasks (#9574)
- Rename ObjectID => ObjectRef (#9353)
- New GCS Actor manager on by default (#8845, #9883, #9715, #9473, #9275)
- Worker towards placement groups (#9039) 
- Plasma store process is merged with raylet (#8939, #8897)
- Option to automatically reconstruct objects stored in plasma after a failure. See the [documentation](https://docs.ray.io/en/master/fault-tolerance.html#objects) for more information. (#9394, #9557, #9488)
- Many bug fixes.

RLlib
-----
- New algorithm: __“Model-Agnostic Meta-Learning” (MAML)__. An algo that learns and generalizes well across a __distribution__ of environments.
- New algorithm: __“Model-Based Meta-Policy-Optimization” (MB-MPO)__. Our first __model-based RL algo__.
- __Windows__ is now __officially supported__ by RLlib.
- __Native TensorFlow 2.x support__. Use framework=”tf2” in your config to tap into TF2’s full potential. Also: SAC, DDPG, DQN Rainbow, ES, and ARS now run in TF1.x Eager mode.
- __DQN PyTorch__ support for full Rainbow setup (including distributional DQN).
- __Python type hints__ for Policy, Model, Offline, Evaluation, and Env classes.
- __Deprecated “Policy Optimizer”__ package (in favor of new distributed execution API).
- Enhanced __test coverage__ and __stability__.
- __Flexible multi-agent replay modes__ and `replay_sequence_length`. We now allow a) storing sequences (over time) in replay buffers and retrieving “lock-stepped” multi-agent samples.
- Environments: __Unity3D soccer game__ (tuned example/benchmark) and __DM Control__ Suite wrapper and examples.
- Various __Bug fixes__: QMIX not learning, DDPG torch bugs, IMPALA learning rate updates, PyTorch custom loss, PPO not learning MuJoCo due to action clipping bug, DQN w/o dueling layer error.

Tune
-----

- **API Changes**:
    - The Tune Function API now supports checkpointing and is now usable with all search and scheduling algorithms! (#8471, #9853, #9517)
    - The Trainable class API has renamed many of its methods to be

_[Truncated at 4000 characters — full notes: https://github.com/ray-project/ray/releases/tag/ray-0.8.7]_