python/ray/data/_internal/arrow_block.py
python/ray/data/_internal/arrow_ops/transform_polars.py
python/ray/data/_internal/arrow_ops/transform_pyarrow.py
python/ray/data/_internal/arrow_utils.py
python/ray/data/_internal/batcher.py
python/ray/data/_internal/block_batching/iter_batches.py
python/ray/data/_internal/block_batching/util.py
python/ray/data/_internal/cluster_autoscaler/default_autoscaling_coordinator.py
python/ray/data/_internal/cluster_autoscaler/default_cluster_autoscaler_v2.py
python/ray/data/_internal/dataset_repr.py
python/ray/data/_internal/datasource/bigquery_datasource.py
python/ray/data/_internal/datasource/binary_datasource.py
python/ray/data/_internal/datasource/clickhouse_datasink.py
python/ray/data/_internal/datasource/csv_datasink.py
python/ray/data/_internal/datasource/csv_datasource.py
python/ray/data/_internal/datasource/delta_sharing_datasource.py
python/ray/data/_internal/datasource/hudi_datasource.py
python/ray/data/_internal/datasource/huggingface_datasource.py
python/ray/data/_internal/datasource/iceberg_datasink.py
python/ray/data/_internal/datasource/iceberg_datasource.py
python/ray/data/_internal/datasource/image_datasink.py
python/ray/data/_internal/datasource/image_datasource.py
python/ray/data/_internal/datasource/json_datasink.py
python/ray/data/_internal/datasource/json_datasource.py
python/ray/data/_internal/datasource/kafka_datasource.py
python/ray/data/_internal/datasource/lance_datasink.py
python/ray/data/_internal/datasource/lance_datasource.py
python/ray/data/_internal/datasource/mongo_datasource.py
python/ray/data/_internal/datasource/numpy_datasink.py
python/ray/data/_internal/datasource/parquet_datasink.py
python/ray/data/_internal/datasource/parquet_datasource.py
python/ray/data/_internal/datasource/range_datasource.py
python/ray/data/_internal/datasource/sql_datasource.py
python/ray/data/_internal/datasource/tfrecords_datasink.py
python/ray/data/_internal/datasource/tfrecords_datasource.py
python/ray/data/_internal/datasource/torch_datasource.py
python/ray/data/_internal/datasource/turbopuffer_datasink.py
python/ray/data/_internal/datasource/video_datasource.py
python/ray/data/_internal/datasource/webdataset_datasink.py
python/ray/data/_internal/datasource/webdataset_datasource.py
python/ray/data/_internal/datasource_v2/listing/file_indexer.py
python/ray/data/_internal/datasource_v2/listing/file_manifest.py
python/ray/data/_internal/datasource_v2/listing/indexing_utils.py
python/ray/data/_internal/equalize.py
python/ray/data/_internal/execution/autoscaling_requester.py
python/ray/data/_internal/execution/backpressure_policy/concurrency_cap_backpressure_policy.py
python/ray/data/_internal/execution/backpressure_policy/downstream_capacity_backpressure_policy.py
python/ray/data/_internal/execution/bundle_queue/fifo.py
python/ray/data/_internal/execution/bundle_queue/hash_link.py
python/ray/data/_internal/execution/bundle_queue/reordering.py
python/ray/data/_internal/execution/callbacks/insert_issue_detectors.py
python/ray/data/_internal/execution/dataset_state.py
python/ray/data/_internal/execution/interfaces/common.py
python/ray/data/_internal/execution/interfaces/execution_options.py
python/ray/data/_internal/execution/interfaces/op_runtime_metrics.py
python/ray/data/_internal/execution/interfaces/physical_operator.py
python/ray/data/_internal/execution/interfaces/ref_bundle.py
python/ray/data/_internal/execution/legacy_compat.py
python/ray/data/_internal/execution/node_trackers/actor_location.py
python/ray/data/_internal/execution/operators/actor_pool_map_operator.py
python/ray/data/_internal/execution/operators/aggregate_num_rows.py
python/ray/data/_internal/execution/operators/base_physical_operator.py
python/ray/data/_internal/execution/operators/hash_aggregate.py
python/ray/data/_internal/execution/operators/hash_shuffle.py
python/ray/data/_internal/execution/operators/input_data_buffer.py
python/ray/data/_internal/execution/operators/join.py
python/ray/data/_internal/execution/operators/limit_operator.py
python/ray/data/_internal/execution/operators/map_operator.py
python/ray/data/_internal/execution/operators/map_transformer.py
python/ray/data/_internal/execution/operators/output_splitter.py
python/ray/data/_internal/execution/operators/task_pool_map_operator.py
python/ray/data/_internal/execution/operators/union_operator.py
python/ray/data/_internal/execution/operators/zip_operator.py
python/ray/data/_internal/execution/ranker.py
python/ray/data/_internal/execution/resource_manager.py
python/ray/data/_internal/execution/streaming_executor.py
python/ray/data/_internal/execution/streaming_executor_state.py
python/ray/data/_internal/execution/util.py
python/ray/data/_internal/issue_detection/detectors/hanging_detector.py
python/ray/data/_internal/issue_detection/issue_detector_configuration.py
python/ray/data/_internal/issue_detection/issue_detector_manager.py
python/ray/data/_internal/iterator/iterator_impl.py
python/ray/data/_internal/iterator/stream_split_iterator.py
python/ray/data/_internal/logging.py
python/ray/data/_internal/logical/interfaces/logical_operator.py
python/ray/data/_internal/logical/operators/from_operators.py
python/ray/data/_internal/logical/operators/input_data_operator.py
python/ray/data/_internal/logical/operators/count_operator.py
python/ray/data/_internal/logical/operators/all_to_all_operator.py
python/ray/data/_internal/logical/operators/join_operator.py
python/ray/data/_internal/logical/operators/map_operator.py
python/ray/data/_internal/logical/operators/n_ary_operator.py
python/ray/data/_internal/logical/operators/one_to_one_operator.py
python/ray/data/_internal/logical/operators/read_operator.py
python/ray/data/_internal/logical/operators/streaming_split_operator.py
python/ray/data/_internal/logical/operators/write_operator.py
python/ray/data/_internal/logical/optimizers.py
python/ray/data/_internal/logical/rules/combine_shuffles.py
python/ray/data/_internal/logical/rules/configure_map_task_memory.py
python/ray/data/_internal/logical/rules/inherit_batch_format.py
python/ray/data/_internal/logical/rules/inherit_target_max_block_size.py
python/ray/data/_internal/logical/rules/limit_pushdown.py
python/ray/data/_internal/logical/rules/operator_fusion.py
python/ray/data/_internal/logical/rules/predicate_pushdown.py
python/ray/data/_internal/logical/rules/projection_pushdown.py
python/ray/data/_internal/logical/rules/common_subexpr_elimination.py
python/ray/data/_internal/logical/rules/set_read_parallelism.py
python/ray/data/_internal/logical/ruleset.py
python/ray/data/_internal/logical/util.py
python/ray/data/_internal/memory_tracing.py
python/ray/data/_internal/metadata_exporter.py
python/ray/data/_internal/object_extensions/pandas.py
python/ray/data/_internal/operator_event_exporter.py
python/ray/data/_internal/operator_schema_exporter.py
python/ray/data/_internal/output_buffer.py
python/ray/data/_internal/pandas_block.py
python/ray/data/_internal/plan.py
python/ray/data/_internal/planner/aggregate.py
python/ray/data/_internal/planner/checkpoint/plan_read_op.py
python/ray/data/_internal/planner/checkpoint/plan_write_op.py
python/ray/data/_internal/planner/exchange/aggregate_task_spec.py
python/ray/data/_internal/planner/exchange/interfaces.py
python/ray/data/_internal/planner/exchange/pull_based_shuffle_task_scheduler.py
python/ray/data/_internal/planner/exchange/push_based_shuffle_task_scheduler.py
python/ray/data/_internal/planner/exchange/shuffle_task_spec.py
python/ray/data/_internal/planner/exchange/sort_task_spec.py
python/ray/data/_internal/planner/exchange/split_repartition_task_scheduler.py
python/ray/data/_internal/planner/plan_all_to_all_op.py
python/ray/data/_internal/planner/plan_download_op.py
python/ray/data/_internal/planner/plan_expression/expression_evaluator.py
python/ray/data/_internal/planner/plan_expression/expression_visitors.py
python/ray/data/_internal/planner/plan_read_op.py
python/ray/data/_internal/planner/plan_udf_map_op.py
python/ray/data/_internal/planner/plan_write_op.py
python/ray/data/_internal/planner/planner.py
python/ray/data/_internal/planner/random_shuffle.py
python/ray/data/_internal/planner/randomize_blocks.py
python/ray/data/_internal/planner/repartition.py
python/ray/data/_internal/planner/sort.py
python/ray/data/_internal/progress/__init__.py
python/ray/data/_internal/progress/base_progress.py
python/ray/data/_internal/progress/rich_progress.py
python/ray/data/_internal/remote_fn.py
python/ray/data/_internal/row.py
python/ray/data/_internal/size_estimator.py
python/ray/data/_internal/split.py
python/ray/data/_internal/stats.py
python/ray/data/_internal/streaming_repartition.py
python/ray/data/_internal/table_block.py
python/ray/data/_internal/tensor_extensions/arrow.py
python/ray/data/_internal/tensor_extensions/pandas.py
python/ray/data/_internal/tensor_extensions/utils.py
python/ray/data/_internal/util.py
python/ray/data/_internal/utils/arrow_utils.py
python/ray/data/_internal/utils/tensorflow_utils.py
python/ray/data/_internal/utils/transform_pyarrow.py
python/ray/data/aggregate.py
python/ray/data/block.py
python/ray/data/checkpoint/checkpoint_filter.py
python/ray/data/checkpoint/interfaces.py
python/ray/data/collate_fn.py
python/ray/data/context.py
python/ray/data/dataset.py
python/ray/data/datasource/datasink.py
python/ray/data/datasource/datasource.py
python/ray/data/datasource/file_based_datasource.py
python/ray/data/datasource/file_datasink.py
python/ray/data/datasource/file_meta_provider.py
python/ray/data/datasource/partitioning.py
python/ray/data/datasource/path_util.py
python/ray/data/datatype.py
python/ray/data/examples/data/video_processing/main.py
python/ray/data/examples/data/video_processing/video_processor.py
python/ray/data/expressions.py
python/ray/data/grouped_data.py
python/ray/data/iterator.py
python/ray/data/namespace_expressions/arr_namespace.py
python/ray/data/namespace_expressions/dt_namespace.py
python/ray/data/namespace_expressions/list_namespace.py
python/ray/data/namespace_expressions/map_namespace.py
python/ray/data/namespace_expressions/string_namespace.py
python/ray/data/namespace_expressions/struct_namespace.py
python/ray/data/preprocessor.py
python/ray/data/preprocessors/chain.py
python/ray/data/preprocessors/concatenator.py
python/ray/data/preprocessors/discretizer.py
python/ray/data/preprocessors/encoder.py
python/ray/data/preprocessors/imputer.py
python/ray/data/preprocessors/scaler.py
python/ray/data/preprocessors/serialization_handlers.py
python/ray/data/preprocessors/torch.py
python/ray/data/preprocessors/utils.py
python/ray/data/preprocessors/vectorizer.py
python/ray/data/random_access_dataset.py
python/ray/data/read_api.py
python/ray/data/stats.py
python/ray/data/tests/block_batching/test_iter_batches.py
python/ray/data/tests/block_batching/test_util.py
python/ray/data/tests/conftest.py
python/ray/data/tests/datasource/test_arrow.py
python/ray/data/tests/datasource/test_bigquery.py
python/ray/data/tests/datasource/test_clickhouse.py
python/ray/data/tests/datasource/test_csv.py
python/ray/data/tests/datasource/test_daft.py
python/ray/data/tests/datasource/test_databricks_credentials.py
python/ray/data/tests/datasource/test_databricks_uc_datasource.py
python/ray/data/tests/datasource/test_delta_sharing.py
python/ray/data/tests/datasource/test_file_based_datasource.py
python/ray/data/tests/datasource/test_file_datasink.py
python/ray/data/tests/datasource/test_huggingface.py
python/ray/data/tests/datasource/test_iceberg.py
python/ray/data/tests/datasource/test_image.py
python/ray/data/tests/datasource/test_numpy.py
python/ray/data/tests/datasource/test_pandas.py
python/ray/data/tests/datasource/test_parquet.py
python/ray/data/tests/datasource/test_snowflake.py
python/ray/data/tests/datasource/test_text.py
python/ray/data/tests/datasource/test_tf.py
python/ray/data/tests/datasource/test_tfrecords.py
python/ray/data/tests/datasource/test_torch.py
python/ray/data/tests/datasource/test_turbopuffer_datasink.py
python/ray/data/tests/datasource/test_uc_datasource.py
python/ray/data/tests/datasource/test_webdataset.py
python/ray/data/tests/expressions/test_arithmetic.py
python/ray/data/tests/expressions/test_boolean.py
python/ray/data/tests/expressions/test_cast.py
python/ray/data/tests/expressions/test_comparison.py
python/ray/data/tests/expressions/test_namespace_arr.py
python/ray/data/tests/expressions/test_namespace_datetime.py
python/ray/data/tests/expressions/test_namespace_list.py
python/ray/data/tests/expressions/test_namespace_map.py
python/ray/data/tests/expressions/test_namespace_string.py
python/ray/data/tests/expressions/test_namespace_struct.py
python/ray/data/tests/expressions/test_predicate.py
python/ray/data/tests/mock_server.py
python/ray/data/tests/preprocessors/test_backwards_compatibility.py
python/ray/data/tests/preprocessors/test_chain.py
python/ray/data/tests/preprocessors/test_concatenator.py
python/ray/data/tests/preprocessors/test_discretizer.py
python/ray/data/tests/preprocessors/test_encoder.py
python/ray/data/tests/preprocessors/test_hasher.py
python/ray/data/tests/preprocessors/test_imputer.py
python/ray/data/tests/preprocessors/test_preprocessors.py
python/ray/data/tests/preprocessors/test_scaler.py
python/ray/data/tests/preprocessors/test_torch.py
python/ray/data/tests/preprocessors/test_transformer.py
python/ray/data/tests/preprocessors/test_vectorizer.py
python/ray/data/tests/test_actor_pool_map_operator.py
python/ray/data/tests/test_agg_e2e.py
python/ray/data/tests/test_arrow_block.py
python/ray/data/tests/test_arrow_serialization.py
python/ray/data/tests/test_auto_parallelism.py
python/ray/data/tests/test_actor_autoscaler.py
python/ray/data/tests/test_autoscaling_coordinator.py
python/ray/data/tests/test_backpressure_e2e.py
python/ray/data/tests/test_backpressure_policies.py
python/ray/data/tests/test_batcher.py
python/ray/data/tests/test_block_ref_bundler.py
python/ray/data/tests/test_bundle_queue.py
python/ray/data/tests/test_checkpoint.py
python/ray/data/tests/test_common_subexpr_elimination.py
python/ray/data/tests/test_consumption.py
python/ray/data/tests/test_context_propagation.py
python/ray/data/tests/test_cudf_batch_format.py
python/ray/data/tests/test_dataset_aggregrations.py
python/ray/data/tests/test_dataset_creation.py
python/ray/data/tests/test_dataset_iter.py
python/ray/data/tests/test_dataset_limits.py
python/ray/data/tests/test_dataset_stats.py
python/ray/data/tests/test_default_cluster_autoscaler_v2.py
python/ray/data/tests/test_dynamic_block_split.py
python/ray/data/tests/test_ecosystem_dask.py
python/ray/data/tests/test_exceptions.py
python/ray/data/tests/test_execution_optimizer_advanced.py
python/ray/data/tests/test_execution_optimizer_basic.py
python/ray/data/tests/test_execution_optimizer_integrations.py
python/ray/data/tests/test_execution_optimizer_limit_pushdown.py
python/ray/data/tests/test_executor_resource_management.py
python/ray/data/tests/test_filter.py
python/ray/data/tests/test_groupby_e2e.py
python/ray/data/tests/test_hash_shuffle.py
python/ray/data/tests/test_hash_shuffle_aggregator.py
python/ray/data/tests/test_idempotency_optimizer_guards.py
python/ray/data/tests/test_hash_shuffle_v2.py
python/ray/data/tests/test_issue_detection.py
python/ray/data/tests/test_issue_detection_manager.py
python/ray/data/tests/test_iterator.py
python/ray/data/tests/test_join.py
python/ray/data/tests/test_jumbo_arrow_block.py
python/ray/data/tests/test_limit_operator.py
python/ray/data/tests/test_logging.py
python/ray/data/tests/test_logging_dataset.py
python/ray/data/tests/test_map.py
python/ray/data/tests/test_map_batches.py
python/ray/data/tests/test_map_operator.py
python/ray/data/tests/test_map_transformer.py
python/ray/data/tests/test_metadata_provider.py
python/ray/data/tests/test_monotonically_increasing_id.py
python/ray/data/tests/test_numpy_support.py
python/ray/data/tests/test_object_gc.py
python/ray/data/tests/test_op_runtime_metrics.py
python/ray/data/tests/test_operator_fusion.py
python/ray/data/tests/test_operator_schema_export.py
python/ray/data/tests/test_operators.py
python/ray/data/tests/test_optimize.py
python/ray/data/tests/test_output_splitter.py
python/ray/data/tests/test_pandas_block.py
python/ray/data/tests/test_partitioning.py
python/ray/data/tests/test_predicate_pushdown.py
python/ray/data/tests/test_progress_manager.py
python/ray/data/tests/test_projection_fusion.py
python/ray/data/tests/test_push_based_shuffle.py
python/ray/data/tests/test_random_e2e.py
python/ray/data/tests/test_ranker.py
python/ray/data/tests/test_ref_bundle.py
python/ray/data/tests/test_reservation_based_resource_allocator.py
python/ray/data/tests/test_resource_manager.py
python/ray/data/tests/test_size_estimation.py
python/ray/data/tests/test_split.py
python/ray/data/tests/test_state_export.py
python/ray/data/tests/test_stats.py
python/ray/data/tests/test_streaming_executor.py
python/ray/data/tests/test_streaming_integration.py
python/ray/data/tests/test_strict_mode.py
python/ray/data/tests/test_tensor.py
python/ray/data/tests/test_tensor_extension.py
python/ray/data/tests/test_torch_iter_batches.py
python/ray/data/tests/test_torch_tensor_utils.py
python/ray/data/tests/test_transform_pyarrow.py
python/ray/data/tests/test_util.py
python/ray/data/tests/test_with_column.py
python/ray/data/tests/unit/expressions/test_arithmetic.py
python/ray/data/tests/unit/expressions/test_conversion.py
python/ray/data/tests/unit/expressions/test_core.py
python/ray/data/tests/unit/expressions/test_namespace_list.py
python/ray/data/tests/unit/test_arrow_block.py
python/ray/data/tests/unit/test_arrow_type_conversion.py
python/ray/data/tests/unit/test_block.py
python/ray/data/tests/unit/test_bundler.py
python/ray/data/tests/unit/test_data_batch_conversion.py
python/ray/data/tests/unit/test_datatype.py
python/ray/data/tests/unit/test_deduping_schema.py
python/ray/data/tests/unit/test_expression_evaluator.py
python/ray/data/tests/unit/test_expression_visitor.py
python/ray/data/tests/unit/test_fifo_bundle_queue.py
python/ray/data/tests/unit/test_object_extension.py
python/ray/data/tests/unit/test_path_util.py
python/ray/data/tests/unit/test_reordering_bundle_queue.py
python/ray/data/tests/unit/test_resource_manager.py
python/ray/data/tests/unit/test_throughput_solver.py
python/ray/data/tests/unit/test_transform_pyarrow.py
python/ray/data/tests/util.py
python/ray/data/util/data_batch_conversion.py
python/ray/data/util/expression_utils.py
python/ray/data/util/torch_utils.py
python/ray/data/tests/test_gpu_shuffle.py
python/ray/data/tests/datasource/test_mongo.py
python/ray/data/tests/datasource/test_kafka.py
python/ray/data/_internal/gpu_shuffle/rapidsmpf_backend.py
python/ray/data/_internal/gpu_shuffle/hash_shuffle.py
python/ray/data/_internal/execution/bundle_queue/base.py
python/ray/data/_internal/datasource_v2/tests/test_file_partitioners.py
python/ray/data/_internal/datasource_v2/tests/test_parquet_datasource_v2.py
python/ray/data/tests/datasource/test_read_parquet_v2.py
python/ray/data/tests/unit/datasource_v2/test_file_chunker.py
python/ray/data/tests/unit/datasource_v2/test_file_indexer.py
python/ray/data/tests/unit/datasource_v2/test_list_files_op.py
