GBM implementation on Legate. The primary goals of legate-boost is to provide a state-of-the-art distributed GBM implementation on Legate, capable of running on CPUs or GPUs at supercomputer scale.
For developers - see contributing
Run with the legate launcher
legate example_script.pyimport cunumeric as cn
import legateboost as lb
X = cn.random.random((1000, 10))
y = cn.random.random(X.shape[0])
model = lb.LBRegressor(verbose=1, n_estimators=100, random_state=0, max_depth=2).fit(
X, y
)legate-boost can learn distributions for continuous data. This is useful in cases where simply predicting the mean does not carry enough information about the training data:
The above example can be found here: examples/probabilistic_regression.
legate-boost can train on datasets that do not fit into memory by splitting the dataset into batches and training the model with partial_fit.
total_estimators = 100
model = lb.LBRegressor(n_estimators=estimators_per_batch)
for i in range(total_estimators // estimators_per_batch):
X_batch, y_batch = train_batches[i % n_batches]
model.partial_fit(
X_batch,
y_batch,
)
The above example can be found here: examples/batch_training.
legate-boost supports tree models, linear models, kernel ridge regression models, custom user models and any combinations of these models.
The following example shows a model combining linear and decision tree base learners on a synthetic dataset.
model = lb.LBRegressor(base_models=(lb.models.Linear(), lb.models.Tree(max_depth=1),), **params).fit(X, y)
The second example shows a model combining kernel ridge regression and decision tree base learners on the wine quality dataset.
model = lb.LBRegressor(base_models=(lb.models.KRR(sigma=0.5), lb.models.Tree(max_depth=5),), **params).fit(X, y)
If you already have cunumeric and legate-core installed, run the following:
pip install \
--no-build-isolation \
--no-deps \
.For more details on customizing the build and setting up a development environment, see contributing.md.