# Ray 1.0.0

- Product: Ray (https://whatsnew.fyi/product/ray)
- Vendor: Anyscale
- Date: 2020-09-30
- Version: 1.0.0
- Original notes: https://github.com/ray-project/ray/releases/tag/ray-1.0.0
- Permalink: https://whatsnew.fyi/product/ray/releases/1.0.0

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'.

---

- **removed** — Deprecated arguments have been removed from ray.init() and ray start commands
- **changed** — Ray Java API is now stable
- **fixed** — Improved detection of Docker CPU limits
- **added** — Support and documentation for Dask-on-Ray and MARS-on-Ray
- **added** — Placement groups for fine-grained control over scheduling decisions
- **added** — Support for multiple instance types in the same cluster
- **added** — Support for specifying GPU/accelerator type in @ray.remote
- **changed** — Improvements to the memory usage tab and machine view in the dashboard
- **added** — Dashboard now supports visualization of actor states
- **added** — Support for Prometheus metrics reporting
- **added** — Two Model-based RL algorithms: MB-MPO and Dreamer
- **added** — Curiosity intrinsic motivation module via RLlib's Exploration API
- **added** — Documentation for the Distributed Execution API
- **removed** — Model(V1) class, Trainer config keys, and some methods/functions have been removed from RLlib
- **added** — DeepMind Control Suite examples
- **deprecated** — Multiple tune.run parameters have been deprecated: ray_auto_init, run_errored_only, global_checkpoint_period, with_server
- **changed** — tune.run upload_dir, sync_to_cloud, sync_to_driver, sync_on_checkpoint parameters have been moved to tune.SyncConfig
- **added** — mode, metric, time_budget parameters for tune.run
- **added** — Search Algorithms now share a uniform API with create_scheduler/create_searcher shim layer
- **added** — Native callbacks for MXNet, Horovod, Keras, XGBoost, and PytorchLightning

#### Ray 1.0

We're happy to announce the release of Ray 1.0, an important step towards the goal of providing a universal API for distributed computing.

To learn more about Ray 1.0, check out our [blog post](https://www.anyscale.com/blog/announcing-ray-1-0) and [whitepaper](https://docs.ray.io/en/master/whitepaper.html).

##### Ray Core
- The ray.init() and `ray start` commands have been cleaned up to remove deprecated arguments
- The Ray Java API is now stable
- Improved detection of Docker CPU limits 
- Add support and documentation for Dask-on-Ray and MARS-on-Ray: https://docs.ray.io/en/master/ray-libraries.html
- Placement groups for fine-grained control over scheduling decisions: https://docs.ray.io/en/latest/placement-group.html.
- New architecture whitepaper: https://docs.ray.io/en/master/whitepaper.html

###### Autoscaler
- Support for multiple instance types in the same cluster: https://docs.ray.io/en/master/cluster/autoscaling.html
- Support for specifying GPU/accelerator type in `@ray.remote`

###### Dashboard & Metrics
- Improvements to the memory usage tab and machine view
- The dashboard now supports visualization of actor states
- Support for Prometheus metrics reporting: https://docs.ray.io/en/latest/ray-metrics.html

##### RLlib
- Two Model-based RL algorithms were added: MB-MPO (“Model-based meta-policy optimization”) and “Dreamer”. Both algos were benchmarked and are performing comparably to the respective papers’ reported results.
- A “Curiosity” (intrinsic motivation) module was added via RLlib’s Exploration API and benchmarked on a sparse-reward Unity3D environment (Pyramids).
- Added documentation for the Distributed Execution API.
- Removed (already soft-deprecated) APIs: Model(V1) class, Trainer config keys, some methods/functions. Where you would see a warning previously when using these, there will be an error thrown now.
- Added DeepMind Control Suite examples.

##### Tune

**Breaking changes:**
-  Multiple tune.run parameters have been deprecated: `ray_auto_init, run_errored_only, global_checkpoint_period, with_server` (#10518)
- `tune.run(upload_dir, sync_to_cloud, sync_to_driver, sync_on_checkpoint` have been moved to `tune.SyncConfig` [[docs](https://docs.ray.io/en/releases-1.0.0/tune/tutorials/tune-distributed.html#syncing)] (#10518)

**New APIs:**
- `mode, metric, time_budget` parameters for tune.run (#10627, #10642)
- Search Algorithms now share a uniform API: (#10621, #10444). You can also use the new `create_scheduler/create_searcher` shim layer to create search algorithms/schedulers via string, reducing boilerplate code (#10456).
- Native callbacks for: [MXNet, Horovod, Keras, XGBoost, PytorchLightning](https://docs.ray.io/en/releases-1.0.0/tune/api_docs/integration.html) (#10533, #10304, #10509, #10502, #10220)
- PBT runs can be replayed with PopulationBasedTrainingReplay scheduler (#9953)
- Search Algorithms are saved/resumed automatically (#9972)
- New Optuna Search Algorithm [docs](https://docs.ray.io/en/releases-1.0.0/tune/api_docs/suggestion.html#optuna-tune-suggest-optuna-optunasearch) (#10044)
- Tune now can sync checkpoints across Kubernetes pods (#10097)
- Failed trials can be rerun with `tune.run(resume="run_errored_only")` (#10060)

**Other Changes:**
- Trial outputs can be saved to file via `tune.run(log_to_file=...)` (#9817)
- Trial directories can be customized, and default trial directory now includes trial name (#10608, #10214)
- Improved Experiment Analysis API (#10645)
- Support for Multi-objective search via SigOpt Wrapper (#10457, #10446)
- BOHB Fixes (#10531, #10320)
- Wandb improvements + RLlib compatibility (#10950, #10799, #10680, #10654, #10614, #10441, #10252, #8521)
- Updated documentation for FAQ, Tune+serve, search space API, lifecycle (#10813, #10925, #10662, #10576, #9713, #10222, #10126, #9908)


##### RaySGD: 
* Creator functions are subsumed by the TrainingOperator API (#10321)
* Training happens

_[Truncated at 4000 characters — full notes: https://github.com/ray-project/ray/releases/tag/ray-1.0.0]_