def _fit_model(population_df, peripheral_df, population_ph, peripheral_ph, seed, units):
# ----------------------------------------------------------------
predictor = predictors.LinearRegression()
# ----------------------------------------------------------------
model = models.MultirelModel(
aggregation=[
aggregations.Count,
aggregations.Sum
],
population=population_ph,
peripheral=[peripheral_ph],
loss_function=loss_functions.SquareLoss(),
predictor=predictor,
num_features=10,
share_aggregations=1.0,
max_length=3,
num_threads=1,
seed=seed,
units=units
).send()
# ----------------------------------------------------------------
model = model.fit(
population_table=population_df,
peripheral_tables=[peripheral_df]
)
# ----------------------------------------------------------------
features = model.transform(
population_table=population_df,
peripheral_tables=[peripheral_df]
)
# ----------------------------------------------------------------
yhat = model.predict(
population_table=population_df,
peripheral_tables=[peripheral_df]
)
# ----------------------------------------------------------------
scores = model.score(
population_table=population_df,
peripheral_tables=[peripheral_df]
)
# ----------------------------------------------------------------
return model, features, yhat, scores
n_jobs=6,
max_depth=7,
reg_lambda=0.0)
#predictor = predictors.LogisticRegression()
model = models.MultirelModel(
population=population_placeholder,
peripheral=[expd_placeholder, memd_placeholder],
loss_function=loss_functions.CrossEntropyLoss(),
aggregation=[
aggregations.Avg, aggregations.Count, aggregations.CountDistinct,
aggregations.CountMinusCountDistinct, aggregations.Max,
aggregations.Median, aggregations.Min, aggregations.Sum,
aggregations.Var
],
num_features=10,
share_aggregations=0.2,
#feature_selector=feature_selector,
predictor=predictor,
allow_sets=True,
max_length=3,
min_num_samples=200,
num_threads=4,
shrinkage=0.1,
include_categorical=True).send()
# -----------------------------------------------------------------------------
# Fit model
model = model.fit(population_table=df_population_training,
peripheral_tables=[df_expd, df_memd])
# ----------------
# Build model
population_placeholder = data.Placeholder(name="numerical_population")
peripheral_placeholder = data.Placeholder(name="numerical_peripheral")
population_placeholder.join(peripheral_placeholder, "join_key", "time_stamp")
predictor = predictors.LinearRegression()
model = models.MultirelModel(
aggregation=[aggregations.Count, aggregations.Sum],
population=population_placeholder,
peripheral=[peripheral_placeholder],
loss_function=loss_functions.SquareLoss(),
predictor=predictor,
num_features=10,
share_aggregations=1.0,
max_length=3,
num_threads=0).send()
# ----------------
model = model.fit(population_table=population_on_engine,
peripheral_tables=[peripheral_on_engine])
# ----------------
features = model.transform(population_table=population_on_engine,
peripheral_tables=[peripheral_on_engine])
# t1.time_stamp;
population_table, peripheral_table = datasets.make_numerical()
population_placeholder = population_table.to_placeholder()
peripheral_placeholder = peripheral_table.to_placeholder()
population_placeholder.join(peripheral_placeholder, "join_key", "time_stamp")
predictor = predictors.LinearRegression()
model = models.MultirelModel(
name="MyModel",
aggregation=[aggregations.Count, aggregations.Sum],
population=population_placeholder,
peripheral=[peripheral_placeholder],
loss_function=loss_functions.SquareLoss(),
predictor=predictor,
include_categorical=True,
num_features=10,
share_aggregations=1.0,
max_length=1,
num_threads=0).send()
# ----------------
model = model.fit(population_table=population_table,
peripheral_tables=[peripheral_table])
# ----------------
model.transform(population_table=population_table,
peripheral_tables=[peripheral_table],
reg_lambda=500)
predictor = predictors.XGBoostClassifier(booster="gbtree",
n_estimators=100,
n_jobs=6,
max_depth=7,
reg_lambda=500)
model = models.MultirelModel(population=population_placeholder,
peripheral=[expd_placeholder, memd_placeholder],
loss_function=loss_functions.CrossEntropyLoss(),
aggregation=[
aggregations.Avg, aggregations.Count,
aggregations.CountDistinct,
aggregations.CountMinusCountDistinct,
aggregations.Max, aggregations.Median,
aggregations.Min, aggregations.Sum,
aggregations.Var
],
feature_selector=feature_selector,
predictor=predictor,
allow_sets=True,
num_threads=3).send()
# ----------------
# Build a hyperparameter space
param_space = dict()
param_space["grid_factor"]: [1.0, 16.0]
param_space["max_length"]: [1, 10]
import getml.loss_functions as loss_functions
model = models.MultirelModel(
population=CE_placeholder,
peripheral=[CE_placeholder],
predictor=predictor,
loss_function=loss_functions.CrossEntropyLoss(),
aggregation=[
aggregations.Avg,
aggregations.Count,
aggregations.CountDistinct,
aggregations.CountMinusCountDistinct,
aggregations.Max,
aggregations.Median,
aggregations.Min,
aggregations.Sum
],
use_timestamps=True,
num_features=70,
max_length=7,
min_num_samples=100,
shrinkage=0.1,
grid_factor=1.0,
regularization=0.0,
round_robin=False,
share_aggregations=0.04,
share_conditions=0.8,
sampling_factor=1.0
).send()
population_placeholder.join(
peripheral_placeholder,
join_key="join_key",
time_stamp="time_stamp_lagged",
other_time_stamp="time_stamp",
)
predictor = predictors.LinearRegression()
model = models.MultirelModel(
aggregation=[aggregations.Count, aggregations.Sum],
population=population_placeholder,
peripheral=[peripheral_placeholder],
loss_function=loss_functions.SquareLoss(),
predictor=predictor,
min_num_samples=1,
num_features=10,
share_aggregations=1.0,
max_length=2,
num_threads=4,
delta_t=1.0 # Define the time delta
).send()
# ----------------
model = model.fit(population_table=population_on_engine,
peripheral_tables=[peripheral_on_engine])
# ----------------
features = model.transform(population_table=population_on_engine,