&`i3 *ddlZddlmZddlmZmZmZmZmZddl Z ddl m Z ddl Z ddl mZddlmZddlmZddlmZdd lmZmZdd lmZdd lmZejeZd Z d e!de"de#de$de!f dZ%edGddeZdS)N)partial)AnyCallableDictOptionalUnion)Version) Checkpoint)TRAIN_DATASET_KEY) GenDataset)_log_deprecation_warning)RayTrainReportCallback XGBoostConfig)XGBoostTrainer) PublicAPIaKPassing in `xgboost.train` kwargs such as `params`, `num_boost_round`, `label_column`, etc. to `XGBoostTrainer` is deprecated in favor of the new API which accepts a training function, similar to the other DataParallelTrainer APIs (ex: TorchTrainer). See this issue for more context: https://github.com/ray-project/ray/issues/50042config label_columnnum_boost_round dataset_keysxgboost_train_kwargsc Ltj}d}|}|rNtj|}|}||z }t d|d|dtjt} | } d|D} d| D} | |d| |}} tj| |}|tfg}| D]O\}}| |d||}}|tj|||fPi}tj|f|||||d |dS) Nz7Model loaded from checkpoint will train for additional zY iterations (trees) in order to achieve the target number of iterations (num_boost_round=z).cdi|]-}|tk |tj|.S)r raytrainget_dataset_shard).0ks u/home/jaya/work/projects/VOICE-AGENT/VIET/agent-env/lib/python3.11/site-packages/ray/train/xgboost/xgboost_trainer.py z0_xgboost_train_fn_per_worker..6s@ ! ! ! 39 & &q ) ) ! ! !cbi|],\}}||-Sr) materialize to_pandas)rrds rr z0_xgboost_train_fn_per_worker..;s2QQQ41a1==??,,..QQQr!)axis)label)dtrainevals evals_resultr xgb_model)rrget_checkpointr get_modelnum_boosted_roundsloggerinforr r#r$itemsdropxgboostDMatrixappend)rrrrr checkpointstarting_modelremaining_iters starting_iter train_ds_itertrain_df eval_ds_iterseval_dfstrain_Xtrain_yr)r* eval_nameeval_dfeval_Xeval_yr+s r_xgboost_train_fn_per_workerrEs))++JN%O  /9*EE&99;; )M9  %) % %  % % %   I//0ABBM((**4466HM RQ=;N;N;P;PQQQH}}\}::H\`. train_loop_config: A configuration ``Dict`` to pass in as an argument to ``train_loop_per_worker``. This is typically used for specifying hyperparameters. xgboost_config: The configuration for setting up the distributed xgboost backend. Defaults to using the "rabit" backend. See :class:`~ray.train.xgboost.XGBoostConfig` for more info. datasets: The Ray Datasets to use for training and validation. dataset_config: The configuration for ingesting the input ``datasets``. By default, all the Ray Datasets are split equally across workers. See :class:`~ray.train.DataConfig` for more details. scaling_config: The configuration for how to scale data parallel training. ``num_workers`` determines how many Python processes are used for training, and ``use_gpu`` determines whether or not each process should use GPUs. See :class:`~ray.train.ScalingConfig` for more info. run_config: The configuration for the execution of the training run. See :class:`~ray.train.RunConfig` for more info. resume_from_checkpoint: A checkpoint to resume training from. This checkpoint can be accessed from within ``train_loop_per_worker`` by calling ``ray.train.get_checkpoint()``. metadata: Dict that should be made available via `ray.train.get_context().get_metadata()` and in `checkpoint.get_metadata()` for checkpoints saved from this Trainer. Must be JSON-serializable. label_column: [Deprecated] Name of the label column. A column with this name must be present in the training dataset. params: [Deprecated] XGBoost training parameters. Refer to `XGBoost documentation `_ for a list of possible parameters. num_boost_round: [Deprecated] Target number of boosting iterations (trees in the model). Note that unlike in ``xgboost.train``, this is the target number of trees, meaning that if you set ``num_boost_round=10`` and pass a model that has already been trained for 5 iterations, it will be trained for 5 iterations more, instead of 10 more. **train_kwargs: [Deprecated] Additional kwargs passed to ``xgboost.train()`` function. TN) train_loop_configxgboost_configscaling_config run_configdatasetsdataset_configresume_from_checkpointmetadatarparamsrtrain_loop_per_workerrIrJrKrLrMrNrOrPrrQrc Xttjtdkrtd|du}|r|| || | |}| pi}n| rt dt tt| |||||||||  dS)Nz1.7.0zm`XGBoostTrainer` requires the `xgboost` version to be >= 1.7.0. Upgrade with: `pip install -U "xgboost>=1.7"`)rrLrrrMzPassing `xgboost.train` kwargs to `XGBoostTrainer` is deprecated. In your training function, you can call `xgboost.train(**kwargs)` with arbitrary arguments. ) rRrIrJrKrLrMrNrOrP) r r4 __version__ ImportError_get_legacy_train_fn_per_workerr *LEGACY_XGBOOST_TRAINER_DEPRECATION_MESSAGEsuperr__init__)selfrRrIrJrKrLrMrNrOrPrrQr train_kwargs legacy_api __class__s rrYzXGBoostTrainer.__init__s( 7& ' '''*:*: : :@  +d2  $($H$H%1%) /! %I%% !!' "     $@>@@    nd##,,"7/))!)#9 - r!rreturnc P|pi}|ts9tdtdt||std|pd}t t |dg}td|D}i}|r&|jj } |jj } | |d<| | nd |d <|s| td i|||d<tt||t|| } | S) zzAXGBoostTrainer._get_legacy_train_fn_per_worker..s>% % =EJx!7 8 8% % % % % % r! frequencyNTcheckpoint_at_end)rrrrr)getr ValueErrorlistkeysr rWanycheckpoint_configcheckpoint_frequencyrhr6rrrEset) rZrrLrMrrrauser_supplied_callbackcallback_kwargsrorhtrain_fn_per_workers rrVz.XGBoostTrainer._get_legacy_train_fn_per_workers>r||-.. 5+55"(--//22 55  G  */R !KLLL),,["== !$% % IR% % % " "   #-#?#T * < N +?OK (&7%B!! / 0& H   3FFoFF G G G,5[)% (%+X!5    #"r!r7c*tj|S)z5Retrieve the XGBoost model stored in this checkpoint.)rr.)clsr7s rr.zXGBoostTrainer.get_model3s &/ ;;;r!rc)__name__ __module__ __qualname____doc___handles_checkpoint_freq_handles_checkpoint_at_endrrrrrrr ScalingConfig RunConfigstrr DataConfigr rintrYrV classmethodr4Boosterr. __classcell__)r]s@rrrTs6ddL $!%  7 -126<@48489=7;-1&*+/)-#7 7 7 ' (2t8$hvt|&<< = 7 $D> 7 !/7 !!897 SY017 4Z017 !!567 !) 47 4S>*7 sm7 c3h(!7 ""##7 7 7 7 7 7 r9#"9#SY019#4Z01 9# sm 9# "# 9# 4&$, 9#9#9#9#v<< <<<[<<<<rs77777777777777%%%%%% 111111((((((444444CCCCCCCCGGGGGG******  8 $ $6+3 333 3  3333l Vd<d<d<d<d<)d<d<d<d<d