Skip to content



DistributedGradientBoostedTreesLearner(label: str, task: Task = generic_learner.Task.CLASSIFICATION, weights: Optional[str] = None, ranking_group: Optional[str] = None, uplift_treatment: Optional[str] = None, features: ColumnDefs = None, include_all_columns: bool = False, max_vocab_count: int = 2000, min_vocab_frequency: int = 5, discretize_numerical_columns: bool = False, num_discretized_numerical_bins: int = 255, max_num_scanned_rows_to_infer_semantic: int = 10000, max_num_scanned_rows_to_compute_statistics: int = 10000, data_spec: Optional[DataSpecification] = None, apply_link_function: Optional[bool] = True, force_numerical_discretization: Optional[bool] = False, max_depth: Optional[int] = 6, max_unique_values_for_discretized_numerical: Optional[int] = 16000, maximum_model_size_in_memory_in_bytes: Optional[float] = -1.0, maximum_training_duration_seconds: Optional[float] = -1.0, min_examples: Optional[int] = 5, num_candidate_attributes: Optional[int] = -1, num_candidate_attributes_ratio: Optional[float] = -1.0, num_trees: Optional[int] = 300, pure_serving_model: Optional[bool] = False, random_seed: Optional[int] = 123456, shrinkage: Optional[float] = 0.1, use_hessian_gain: Optional[bool] = False, worker_logs: Optional[bool] = True, num_threads: Optional[int] = None, working_dir: Optional[str] = None, resume_training: bool = False, resume_training_snapshot_interval_seconds: int = 1800, tuner: Optional[AbstractTuner] = None, workers: Optional[Sequence[str]] = None)

Bases: GenericLearner

Distributed Gradient Boosted Trees learning algorithm.

Exact distributed version of the Gradient Boosted Tree learning algorithm. See the documentation of the non-distributed Gradient Boosted Tree learning algorithm for an introduction to GBTs.

Usage example:

import ydf
import pandas as pd

dataset = pd.read_csv("project/dataset.csv")

model = ydf.DistributedGradientBoostedTreesLearner().train(dataset)


Hyperparameters are configured to give reasonable results for typical datasets. Hyperparameters can also be modified manually (see descriptions) below or by applying the hyperparameter templates available with DistributedGradientBoostedTreesLearner.hyperparameter_templates() (see this function's documentation for details).


Name Type Description

Label of the dataset. The label column should not be identified as a feature in the features parameter.




Name of a feature that identifies the weight of each example. If weights are not specified, unit weights are assumed. The weight column should not be identified as a feature in the features parameter.


Only for task=Task.RANKING. Name of a feature that identifies queries in a query/document ranking task. The ranking group should not be identified as a feature in the features parameter.


Only for task=Task.CATEGORICAL_UPLIFT and task=Task. NUMERICAL_UPLIFT. Name of a numerical feature that identifies the treatment in an uplift problem. The value 0 is reserved for the control treatment. Currently, only 0/1 binary treatments are supported.


If None, all columns are used as features. The semantic of the features is determined automatically. Otherwise, if include_all_columns=False (default) only the column listed in features are imported. If include_all_columns=True, all the columns are imported as features and only the semantic of the columns NOT in columns is determined automatically. If specified, defines the order of the features - any non-listed features are appended in-order after the specified features (if include_all_columns=True). The label, weights, uplift treatment and ranking_group columns should not be specified as features.


See features.


Maximum size of the vocabulary of CATEGORICAL and CATEGORICAL_SET columns stored as strings. If more unique values exist, only the most frequent values are kept, and the remaining values are considered as out-of-vocabulary.


Minimum number of occurrence of a value for CATEGORICAL and CATEGORICAL_SET columns. Value observed less than min_vocab_frequency are considered as out-of-vocabulary.


If true, discretize all the numerical columns before training. Discretized numerical columns are faster to train with, but they can have a negative impact on the model quality. Using discretize_numerical_columns=True is equivalent as setting the column semantic DISCRETIZED_NUMERICAL in the column argument. See the definition of DISCRETIZED_NUMERICAL for more details.


Number of bins used when disretizing numerical columns.


Number of rows to scan when inferring the column's semantic if it is not explicitly specified. Only used when reading from file, in-memory datasets are always read in full. Setting this to a lower number will speed up dataset reading, but might result in incorrect column semantics. Set to -1 to scan the entire dataset.


Number of rows to scan when computing a column's statistics. Only used when reading from file, in-memory datasets are always read in full. A column's statistics include the dictionary for categorical features and the mean / min / max for numerical features. Setting this to a lower number will speed up dataset reading, but skew statistics in the dataspec, which can hurt model quality (e.g. if an important category of a categorical feature is considered OOV). Set to -1 to scan the entire dataset.


Dataspec to be used (advanced). If a data spec is given, columns, include_all_columns, max_vocab_count, min_vocab_frequency, discretize_numerical_columns and num_discretized_numerical_bins will be ignored.


If true, applies the link function (a.k.a. activation function), if any, before returning the model prediction. If false, returns the pre-link function model output. For example, in the case of binary classification, the pre-link function output is a logic while the post-link function is a probability. Default: True.


If false, only the numerical column safisfying "max_unique_values_for_discretized_numerical" will be discretized. If true, all the numerical columns will be discretized. Columns with more than "max_unique_values_for_discretized_numerical" unique values will be approximated with "max_unique_values_for_discretized_numerical" bins. This parameter will impact the model training. Default: False.


Maximum depth of the tree. max_depth=1 means that all trees will be roots. max_depth=-1 means that tree depth is not restricted by this parameter. Values <= -2 will be ignored. Default: 6.


Maximum number of unique value of a numerical feature to allow its pre-discretization. In case of large datasets, discretized numerical features with a small number of unique values are more efficient to learn than classical / non-discretized numerical features. This parameter does not impact the final model. However, it can speed-up or slown the training. Default: 16000.


Limit the size of the model when stored in ram. Different algorithms can enforce this limit differently. Note that when models are compiled into an inference, the size of the inference engine is generally much smaller than the original model. Default: -1.0.


Maximum training duration of the model expressed in seconds. Each learning algorithm is free to use this parameter at it sees fit. Enabling maximum training duration makes the model training non-deterministic. Default: -1.0.


Minimum number of examples in a node. Default: 5.


Number of unique valid attributes tested for each node. An attribute is valid if it has at least a valid split. If num_candidate_attributes=0, the value is set to the classical default value for Random Forest: sqrt(number of input attributes) in case of classification and number_of_input_attributes / 3 in case of regression. If num_candidate_attributes=-1, all the attributes are tested. Default: -1.


Ratio of attributes tested at each node. If set, it is equivalent to num_candidate_attributes = number_of_input_features x num_candidate_attributes_ratio. The possible values are between ]0, and 1] as well as -1. If not set or equal to -1, the num_candidate_attributes is used. Default: -1.0.


Maximum number of decision trees. The effective number of trained tree can be smaller if early stopping is enabled. Default: 300.


Clear the model from any information that is not required for model serving. This includes debugging, model interpretation and other meta-data. The size of the serialized model can be reduced significatively (50% model size reduction is common). This parameter has no impact on the quality, serving speed or RAM usage of model serving. Default: False.


Random seed for the training of the model. Learners are expected to be deterministic by the random seed. Default: 123456.


Coefficient applied to each tree prediction. A small value (0.02) tends to give more accurate results (assuming enough trees are trained), but results in larger models. Analogous to neural network learning rate. Default: 0.1.


Use true, uses a formulation of split gain with a hessian term i.e. optimizes the splits to minimize the variance of "gradient / hessian. Available for all losses except regression. Default: False.


If true, workers will print training logs. Default: True.


Number of threads used to train the model. Different learning algorithms use multi-threading differently and with different degree of efficiency. If None, num_threads will be automatically set to the number of processors (up to a maximum of 32; or set to 6 if the number of processors is not available). Making num_threads significantly larger than the number of processors can slow-down the training speed. The default value logic might change in the future.


If true, the model training resumes from the checkpoint stored in the working_dir directory. If working_dir does not contain any model checkpoint, the training starts from the beginning. Resuming training is useful in the following situations: (1) The training was interrupted by the user (e.g. ctrl+c or "stop" button in a notebook) or rescheduled, or (2) the hyper-parameter of the learner was changed e.g. increasing the number of trees.


Path to a directory available for the learning algorithm to store intermediate computation results. Depending on the learning algorithm and parameters, the working_dir might be optional, required, or ignored. For instance, distributed training algorithm always need a "working_dir", and the gradient boosted tree and hyper-parameter tuners will export artefacts to the "working_dir" if provided.


Indicative number of seconds in between snapshots when resume_training=True. Might be ignored by some learners.


If set, automatically select the best hyperparameters using the provided tuner. When using distributed training, the tuning is distributed.


If set, enable distributed training. "workers" is the list of IP addresses of the workers. A worker is a process running ydf.start_worker(port).


cross_validation(ds: InputDataset, folds: int = 10, bootstrapping: Union[bool, int] = False, parallel_evaluations: int = 1) -> Evaluation

Cross-validates the learner and return the evaluation.

Usage example:

import pandas as pd
import ydf

dataset = pd.read_csv("my_dataset.csv")
learner = ydf.RandomForestLearner(label="label")
evaluation = learner.cross_validation(dataset)

# In a notebook, display an interractive evaluation

# Print the evaluation

# Look at specific metrics


Name Type Description Default
ds InputDataset

Dataset for the cross-validation.

folds int

Number of cross-validation folds.

bootstrapping Union[bool, int]

Controls whether bootstrapping is used to evaluate the confidence intervals and statistical tests (i.e., all the metrics ending with "[B]"). If set to false, bootstrapping is disabled. If set to true, bootstrapping is enabled and 2000 bootstrapping samples are used. If set to an integer, it specifies the number of bootstrapping samples to use. In this case, if the number is less than 100, an error is raised as bootstrapping will not yield useful results.

parallel_evaluations int

Number of model to train and evaluate in parallel using multi-threading. Note that each model is potentially already trained with multithreading (see num_threads argument of Learner constructor).



Type Description

The cross-validation evaluation.

hyperparameter_templates classmethod

hyperparameter_templates() -> Dict[str, HyperparameterTemplate]

Hyperparameter templates for this Learner.

This learner currently does not provide any hyperparameter templates, this method is provided for consistency with other learners.


Type Description
Dict[str, HyperparameterTemplate]

Empty dictionary.


train(ds: InputDataset, valid: Optional[InputDataset] = None) -> GradientBoostedTreesModel

Trains a model on the given dataset.

Options for dataset reading are given on the learner. Consult the documentation of the learner or ydf.create_vertical_dataset() for additional information on dataset reading in YDF.

Usage example:

import ydf
import pandas as pd

train_ds = pd.read_csv(...)

learner = ydf.DistributedGradientBoostedTreesLearner(label="label")
model = learner.train(train_ds)

If training is interrupted (for example, by interrupting the cell execution in Colab), the model will be returned to the state it was in at the moment of interruption.


Name Type Description Default
ds InputDataset

Training dataset.

valid Optional[InputDataset]

Optional validation dataset. Some learners, such as Random Forest, do not need validation dataset. Some learners, such as GradientBoostedTrees, automatically extract a validation dataset from the training dataset if the validation dataset is not provided.



Type Description

A trained model.