Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Update Ray Dataset with prediction example #287

Merged
merged 10 commits into from
Aug 4, 2023
Merged

Update Ray Dataset with prediction example #287

Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
b2bf877
Update Ray Dataset with prediction example
Jul 24, 2023
36691f5
Update CI to use Python 3.8-3.10
Aug 3, 2023
b0be1ae
string
Aug 3, 2023
69b6a05
gpu
Aug 3, 2023
f0dcd62
qid
Aug 3, 2023
1cc41da
monotonic
Aug 3, 2023
7a07d2c
pandas 2.0
Aug 4, 2023
56e9162
Merge branch 'ci-py38' into ray-data-pred
Aug 4, 2023
df8968d
Merge branch 'master' into ray-data-pred
Aug 4, 2023
2fbaccb
no repartition
Aug 4, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 9 additions & 10 deletions xgboost_ray/examples/simple_ray_dataset.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,11 +3,13 @@
import numpy as np
import pandas as pd
import ray
from xgboost import DMatrix

from xgboost_ray import RayDMatrix, RayParams, train


def main(cpus_per_actor, num_actors):
np.random.seed(1234)
# Generate dataset
x = np.repeat(range(8), 16).reshape((32, 4))
# Even numbers --> 0, odd numbers --> 1
Expand All @@ -22,16 +24,7 @@ def main(cpus_per_actor, num_actors):
data.columns = [str(c) for c in data.columns]
data["label"] = y

# There was recent API change - the first clause covers the new
# and current Ray master API
if hasattr(ray.data, "from_pandas_refs"):
# Generate Ray dataset from 4 partitions
ray_ds = ray.data.from_pandas(data).repartition(num_actors)
else:
# Split into 4 partitions
partitions = [ray.put(part) for part in np.split(data, num_actors)]
ray_ds = ray.data.from_pandas(partitions)

ray_ds = ray.data.from_pandas(data)
train_set = RayDMatrix(ray_ds, "label")

evals_result = {}
Expand Down Expand Up @@ -62,6 +55,12 @@ def main(cpus_per_actor, num_actors):
bst.save_model(model_path)
print("Final training error: {:.4f}".format(evals_result["train"]["error"][-1]))

# Distributed prediction
scored = ray_ds.drop_columns(["label"]).map_batches(
lambda batch: {"pred": bst.predict(DMatrix(batch))}, batch_format="pandas"
)
print(scored.to_pandas())


if __name__ == "__main__":
parser = argparse.ArgumentParser()
Expand Down
Loading