RelboostModel

class getml.feature_learning.RelboostModel(allow_null_weights: bool = False, delta_t: float = 0.0, gamma: float = 0.0, loss_function: str = 'SquareLoss', max_depth: int = 3, min_df: int = 30, min_num_samples: int = 1, num_features: int = 100, num_subfeatures: int = 100, num_threads: int = 0, propositionalization: getml.feature_learning.fastprop_model.FastPropModel = FastPropModel(aggregation=['AVG', 'COUNT', 'COUNT DISTINCT', 'COUNT MINUS COUNT DISTINCT', 'FIRST', 'LAST', 'MAX', 'MEDIAN', 'MIN', 'MODE', 'STDDEV', 'SUM'], loss_function='SquareLoss', min_df=30, n_most_frequent=0, num_features=200, num_threads=0, sampling_factor=1.0, silent=True, vocab_size=500), reg_lambda: float = 0.0, sampling_factor: float = 1.0, seed: int = 5543, shrinkage: float = 0.1, silent: bool = True, use_timestamps: bool = True, vocab_size: int = 500)

Feature learning based on Gradient Boosting.

RelboostModel automates feature learning for relational data and time series. It is based on a generalization of the XGBoost algorithm to relational data, hence the name.

For more information on the underlying feature learning algorithm, check out Relboost.

Args:

allow_null_weights (bool, optional):

Whether you want to allow RelboostModel to set weights to NULL.

delta_t (float, optional):

Frequency with which lag variables will be explored in a time series setting. When set to 0.0, there will be no lag variables.

For more information, please refer to Time series. Range: [0, \(\infty\)]

gamma (float, optional):

During the training of Relboost, which is based on gradient tree boosting, this value serves as the minimum improvement in terms of the loss_function required for a split of the tree to be applied. Larger gamma will lead to fewer partitions of the tree and a more conservative algorithm. Range: [0, \(\infty\)]

loss_function (loss_functions, optional):

Objective function used by the feature learning algorithm to optimize your features. For regression problems use SquareLoss and for classification problems use CrossEntropyLoss.

max_depth (int, optional):

Maximum depth of the trees generated during the gradient tree boosting. Deeper trees will result in more complex models and increase the risk of overfitting. Range: [0, \(\infty\)]

min_df (int, optional):

Only relevant for columns with role text. The minimum number of fields (i.e. rows) in text column a given word is required to appear in to be included in the bag of words. Range: [1, \(\infty\)]

min_num_samples (int, optional):

Determines the minimum number of samples a subcondition should apply to in order for it to be considered. Higher values lead to less complex statements and less danger of overfitting. Range: [1, \(\infty\)]

num_features (int, optional):

Number of features generated by the feature learning algorithm. Range: [1, \(\infty\)]

num_subfeatures (int, optional):

The number of subfeatures you would like to extract in a subensemble (for snowflake data model only). See The snowflake schema for more information. Range: [1, \(\infty\)]

num_threads (int, optional):

Number of threads used by the feature learning algorithm. If set to zero or a negative value, the number of threads will be determined automatically by the getML engine. Range: [\(0\), \(\infty\)]

propositionalization (FastPropModel, optional):

The feature learner used for joins, which are flagged to be propositionalized (through setting a join’s relationship parameter to getml.data.relationship.propositionalization)

reg_lambda (float, optional):

L2 regularization on the weights in the gradient boosting routine. This is one of the most important hyperparameters in the RelboostModel as it allows for the most direct regularization. Larger values will make the resulting model more conservative. Range: [0, \(\infty\)]

sampling_factor (float, optional):

Relboost uses a bootstrapping procedure (sampling with replacement) to train each of the features. The sampling factor is proportional to the share of the samples randomly drawn from the population table every time Relboost generates a new feature. A lower sampling factor (but still greater than 0.0), will lead to less danger of overfitting, less complex statements and faster training. When set to 1.0, roughly 20,000 samples are drawn from the population table. If the population table contains less than 20,000 samples, it will use standard bagging. When set to 0.0, there will be no sampling at all. Range: [0, \(\infty\)]

seed (Union[int,None], optional):

Seed used for the random number generator that underlies the sampling procedure to make the calculation reproducible. Internally, a seed of None will be mapped to 5543. Range: [0, \(\infty\)]

shrinkage (float, optional):

Since Relboost works using a gradient-boosting-like algorithm, shrinkage (or learning rate) scales down the weights and thus the impact of each new tree. This gives more room for future ones to improve the overall performance of the model in this greedy algorithm. It must be between 0.0 and 1.0 with higher values leading to more danger of overfitting. Range: [0, 1]

silent (bool, optional):

Controls the logging during training.

use_timestamps (bool, optional):

Whether you want to ignore all elements in the peripheral tables where the time stamp is greater than the one in the corresponding elements of the population table. In other words, this determines whether you want add the condition

t2.time_stamp <= t1.time_stamp

at the very end of each feature. It is strongly recommend to enable this behavior.

vocab_size (int, optional):

Determines the maximum number of words that are extracted in total from getml.data.roles.text columns. This can be interpreted as the maximum size of the bag of words. Range: [0, \(\infty\)]

Example:

population_placeholder = getml.data.Placeholder("population")
order_placeholder = getml.data.Placeholder("order")
trans_placeholder = getml.data.Placeholder("trans")

population_placeholder.join(order_placeholder,
                            join_key="account_id")

population_placeholder.join(trans_placeholder,
                            join_key="account_id",
                            time_stamp="date")

feature_selector = getml.predictors.XGBoostClassifier(
    reg_lambda=500
)

predictor = getml.predictors.XGBoostClassifier(
    reg_lambda=500
)

feature_learner = getml.feature_learning.RelboostModel(
    num_features=60,
    loss_function=getml.feature_learning.loss_functions.CrossEntropyLoss
)

pipe = getml.pipeline.Pipeline(
    tags=["relboost", "31 features"],
    population=population_placeholder,
    peripheral=[order_placeholder, trans_placeholder],
    feature_learners=feature_learner,
    feature_selectors=feature_selector,
    predictors=predictor,
    share_selected_features=0.5
)

pipe.check(
    population_table=population_train,
    peripheral_tables={"order": order, "trans": trans}
)

pipe = pipe.fit(
    population_table=population_train,
    peripheral_tables={"order": order, "trans": trans}
)

in_sample = pipe.score(
    population_table=population_train,
    peripheral_tables={"order": order, "trans": trans}
)

out_of_sample = pipe.score(
    population_table=population_test,
    peripheral_tables={"order": order, "trans": trans}
)

Methods

validate([params])

Checks both the types and the values of all instance variables and raises an exception if something is off.

Attributes

allow_null_weights
delta_t
gamma
loss_function
max_depth
min_df
min_num_samples
num_features
num_subfeatures
num_threads
propositionalization
reg_lambda
sampling_factor
seed
shrinkage
silent
type
use_timestamps
vocab_size