Updates TensorflowPredictor to use the new _predict_pandas API.
Also as agreed upon offline, removes the extra configurations from TensorflowPredictor (column selection, concatenation) in favor of having this be done via a Preprocessor.
Update documentation to use `session.report`.
Next steps:
1. Update our internal caller to use `session.report`. Most importantly, CheckpointManager and DataParallelTrainer.
2. Update `get_trial_resources` to use PGF notions to incorporate the requirement of ResourceChangingScheduler. @Yard1
3. After 2 is done, change all `tune.get_trial_resources` to `session.get_trial_resources`
4. [internal implementation] remove special checkpoint handling logic from huggingface trainer. Optimize the flow for checkpoint conversion with `session.report`.
Co-authored-by: Antoni Baum <antoni.baum@protonmail.com>
1. Update `DummyTrainer` to take `num_epochs` instead of `runtime_seconds`.
1. Ray Train expects equal number of calls to `train.report()`. Different workers may run at different speeds and terminate after different epoch numbers, which causes an error.
2. Add `generate_epochs` to support `DatasetPipeline` when `use_stream_api` is True.
3. Update `__main__` code to support testing different configurations.
This PR:
* Allows the user to set `keep_checkpoints_num` and `checkpoint_score_attr` in `RunConfig` using the `CheckpointStrategy` dataclass
* Adds two new fields to the `Result` object - `best_checkpoints` - a list of saved best checkpoints as determined by `CheckpointingConfig`.
As the integration logging callbacks are commonly used with AIR Trainers, they should be moved from the tune package to the air package. The old imports will still work, but raise a deprecation warning.
Follow up on our last discussion for supporting piecemeal fashion air users.
Only did for tensorflow for now, want to collect some feedback on API naming, package structure etc and I will add others.
This adds the following options to DatasetConfig, which can be used to enable streaming ingest.
```
# Whether the dataset should be streamed into memory using pipelined reads.
# When enabled, get_dataset_shard() returns DatasetPipeline instead of Dataset.
# The amount of memory to use is controlled by `stream_window_size`.
# False by default for all datasets.
use_stream_api: Optional[bool] = None
# Configure the streaming window size in bytes. A typical value is something like
# 20% of object store memory. If set to -1, then an infinite window size will be
# used (similar to bulk ingest). This only has an effect if use_stream_api is set.
# Set to 1.0 GiB by default.
stream_window_size: Optional[float] = None
# Whether to enable global shuffle (per pipeline window in streaming mode). Note
# that this is an expensive all-to-all operation, and most likely you want to use
# local shuffle instead.
# False by default for all datasets.
global_shuffle: Optional[bool] = None
```
This adds a per-dataset config object to DataParallelTrainer. These configs define how the Dataset should be read into the DataParallelTrainer. It configures the preprocessing, splitting, and ingest strategy per-dataset. DataParallelTrainers declare default DatasetConfigs for each dataset passed in the ``datasets`` argument. Users have the opportunity to selectively override these configs by passing the ``dataset_config`` argument. Trainers can also define user customizable values (e.g., XGBoostTrainer doesn't support streaming ingest).
This PR adds the minimal support for dataset configs. Future PRs will:
- Add support for streaming ingest
- Move this config from DataParallelTrainer to ml.Trainer
The package "ml" should be renamed to "air".
Main question: Keep a `ml.py` with `from ray.air import *` for some level of backwards compatibility?
I'd go for no to force people to use the new structure.
Currently, we are not running doc notebooks in CI due to a bazel misconfiguration - we are using `glob` in a top level package in order to get the paths for the notebooks, but those are contained inside subpackages, which glob purposefully ignores. Therefore, the lists of notebooks to run are empty. This PR fixes that by:
* Running the `py_test_run_all_notebooks` macro inside the relevant subpackages
* Editing the `test_myst_doc.py` script to allow for recursive search for the target file, allowing to deal with mismatches between `name` and `data` arguments in `py_test_run_all_notebooks`
* Setting the `allow_empty=False` flag inside `glob` calls in our macros to ensure that this oversight is caught early
* Enabling detection of changes in doc folder for `*.ipynb` and `BUILD` files
This PR also adds a GPU runner for doc tests, allowing one of our examples to pass - and setting the infra for more to come. Finally, a misconfigured path for one set of doc tests is also fixed.
Adds a Dataset.split_proportionately method that allows the user to split a dataset using proportions. This is a very common use-case for eg. train-test splitting. The implementation is a thin wrapper over Dataset.split_at_indices.
Additionally, this PR adds a ray.ml.train_test_split function intended to provide a familiar API to ML practitioners.
This is a notebook showing how to tune an xgboost model and analyze the results.
Also adds a `get_dataframe()` method to `ResultsGrid` to fetch the trial results.
Depends on #24483 for toctree.
