def test_tune(ray_start_4_cpus):
train_dataset = ray.data.from_pandas(train_df)
valid_dataset = ray.data.from_pandas(test_df)
trainer = LightGBMTrainer(
scaling_config=scale_config,
label_column="target",
params={
**params,
**{
"max_depth": 1
}
},
datasets={
TRAIN_DATASET_KEY: train_dataset,
"valid": valid_dataset
},
)
tune.run(
trainer.as_trainable(),
config={"params": {
"max_depth": tune.randint(2, 4)
}},
num_samples=2,
)
# Make sure original Trainer is not affected.
assert trainer.params["max_depth"] == 1
def train_lightgbm(num_workers: int, use_gpu: bool = False) -> Result:
train_dataset, valid_dataset, _ = prepare_data()
# Scale some random columns
columns_to_scale = ["mean radius", "mean texture"]
preprocessor = StandardScaler(columns=columns_to_scale)
# LightGBM specific params
params = {
"objective": "binary",
"metric": ["binary_logloss", "binary_error"],
}
trainer = LightGBMTrainer(
scaling_config={
"num_workers": num_workers,
"use_gpu": use_gpu,
},
label_column="target",
params=params,
datasets={"train": train_dataset, "valid": valid_dataset},
preprocessor=preprocessor,
num_boost_round=100,
)
result = trainer.fit()
print(result.metrics)
return result
def test_preprocessor_in_checkpoint(ray_start_4_cpus, tmpdir):
train_dataset = ray.data.from_pandas(train_df)
valid_dataset = ray.data.from_pandas(test_df)
class DummyPreprocessor(Preprocessor):
def __init__(self):
super().__init__()
self.is_same = True
def fit(self, dataset):
self.fitted_ = True
def _transform_pandas(self, df: "pd.DataFrame") -> "pd.DataFrame":
return df
trainer = LightGBMTrainer(
scaling_config=scale_config,
label_column="target",
params=params,
datasets={TRAIN_DATASET_KEY: train_dataset, "valid": valid_dataset},
preprocessor=DummyPreprocessor(),
)
result = trainer.fit()
# Move checkpoint to a different directory.
checkpoint_dict = result.checkpoint.to_dict()
checkpoint = Checkpoint.from_dict(checkpoint_dict)
checkpoint_path = checkpoint.to_directory(tmpdir)
resume_from = Checkpoint.from_directory(checkpoint_path)
model, preprocessor = load_checkpoint(resume_from)
assert get_num_trees(model) == 10
assert preprocessor.is_same
assert preprocessor.fitted_
def test_fit(ray_start_4_cpus):
train_dataset = ray.data.from_pandas(train_df)
valid_dataset = ray.data.from_pandas(test_df)
trainer = LightGBMTrainer(
scaling_config=scale_config,
label_column="target",
params=params,
datasets={
TRAIN_DATASET_KEY: train_dataset,
"valid": valid_dataset
},
)
trainer.fit()
def test_validation(ray_start_4_cpus):
train_dataset = ray.data.from_pandas(train_df)
valid_dataset = ray.data.from_pandas(test_df)
with pytest.raises(KeyError, match=TRAIN_DATASET_KEY):
LightGBMTrainer(
scaling_config={"num_workers": 2},
label_column="target",
params=params,
datasets={"valid": valid_dataset},
)
with pytest.raises(KeyError, match="dmatrix_params"):
LightGBMTrainer(
scaling_config={"num_workers": 2},
label_column="target",
params=params,
dmatrix_params={"data": {}},
datasets={TRAIN_DATASET_KEY: train_dataset, "valid": valid_dataset},
)
def test_resume_from_checkpoint(ray_start_4_cpus, tmpdir):
train_dataset = ray.data.from_pandas(train_df)
valid_dataset = ray.data.from_pandas(test_df)
trainer = LightGBMTrainer(
scaling_config=scale_config,
label_column="target",
params=params,
num_boost_round=5,
datasets={TRAIN_DATASET_KEY: train_dataset, "valid": valid_dataset},
)
result = trainer.fit()
checkpoint = result.checkpoint
model, _ = load_checkpoint(checkpoint)
assert get_num_trees(model) == 5
# Move checkpoint to a different directory.
checkpoint_dict = result.checkpoint.to_dict()
checkpoint = Checkpoint.from_dict(checkpoint_dict)
checkpoint_path = checkpoint.to_directory(tmpdir)
resume_from = Checkpoint.from_directory(checkpoint_path)
trainer = LightGBMTrainer(
scaling_config=scale_config,
label_column="target",
params=params,
num_boost_round=5,
datasets={TRAIN_DATASET_KEY: train_dataset, "valid": valid_dataset},
resume_from_checkpoint=resume_from,
)
result = trainer.fit()
checkpoint = result.checkpoint
xgb_model, _ = load_checkpoint(checkpoint)
assert get_num_trees(xgb_model) == 10
def test_fit_with_categoricals(ray_start_4_cpus):
train_df_with_cat = train_df.copy()
test_df_with_cat = test_df.copy()
train_df_with_cat["categorical_column"] = pd.Series(
(["A", "B"] * math.ceil(len(train_df_with_cat) / 2))[: len(train_df_with_cat)]
).astype("category")
test_df_with_cat["categorical_column"] = pd.Series(
(["A", "B"] * math.ceil(len(test_df_with_cat) / 2))[: len(test_df_with_cat)]
).astype("category")
train_dataset = ray.data.from_pandas(train_df_with_cat)
valid_dataset = ray.data.from_pandas(test_df_with_cat)
trainer = LightGBMTrainer(
scaling_config=scale_config,
label_column="target",
params=params,
datasets={TRAIN_DATASET_KEY: train_dataset, "valid": valid_dataset},
)
result = trainer.fit()
checkpoint = result.checkpoint
model, _ = load_checkpoint(checkpoint)
assert model.pandas_categorical == [["A", "B"]]