betatim

"""
Pipeline(StandardScaler, PCA, RandomForest) with GridSearchCV
=============================================================

GridSearchCV over an all-proxy Pipeline on the full Forest Cover Type
dataset (581K samples, 54 features, 7 classes).

Pipeline: StandardScaler -> PCA -> RandomForestClassifier

All three steps are cuml.accel proxies, so the GridSearchCV patch

betatim

"""
Benchmark: numpy-to-cupy (CPU-to-GPU) transfer times on this machine.

Target: GPU 1 (NVIDIA RTX A6000, 48 GB, PCIe Gen4 x16 slot)
"""

import time
import statistics
import numpy as np
import cupy as cp

betatim

import tarfile
import time
import urllib.request
from collections import OrderedDict
from pathlib import Path

import numpy as np
import scipy.io
import scipy.sparse
from sklearn.decomposition import TruncatedSVD

betatim

AGENTS Instruction

This project is called foobar. Its goal is to provide ...

This file contains is additional guidance for AI agents and other AI editors.

---

Core Principles

betatim

"""
Test: Validate that cuml native estimators can be converted to ONNX
via as_sklearn() -> skl2onnx -> onnxruntime.

Unlike cuml.accel proxies (which skl2onnx recognizes directly), native cuml
estimators must first be converted to sklearn via as_sklearn() before
skl2onnx.convert_sklearn() will accept them.

Run without cuml.accel:
    python test_onnx_as_sklearn.py

betatim

Array API Architecture

Created: 2026-01-07 Last Updated: 2026-01-07

Overview

Scikit-learn's Array API support enables estimators and functions to work with arrays from different libraries (NumPy, CuPy, PyTorch) without modification. This allows computations to run on GPUs when using GPU-backed array libraries.

The implementation follows the Array API Standard, a specification that defines a common API for array manipulation libraries.

betatim

#!/usr/bin/env python3
"""
Ray + RandomForestClassifier with max_calls=1

Demonstrates the impact of max_calls=1 on Ray task execution when using
scikit-learn's RandomForestClassifier.
"""
import time
import ray
from sklearn.datasets import make_classification

betatim

"""
Benchmark: scikit-learn RandomForest vs LightGBM RandomForest

Compares performance across:
- Number of samples (1K, 10K, 100K, 500K)
- Number of features (10, 50, 200)
- Feature types (numerical, categorical, mixed)
- Number of classes (2, 5, 10)

Includes cases optimized for LightGBM's strengths:

betatim

name: tabareana-20251202
channels:
  - conda-forge
dependencies:
  - _libgcc_mutex=0.1=conda_forge
  - _openmp_mutex=4.5=2_gnu
  - bzip2=1.0.8=hda65f42_8
  - ca-certificates=2025.11.12=hbd8a1cb_0
  - cuda-cccl_linux-64=13.0.85=ha770c72_0
  - cuda-cudart-dev_linux-64=13.0.96=h376f20c_0

betatim

from __future__ import annotations

import warnings
warnings.simplefilter("error", FutureWarning)
from pathlib import Path
from typing import Any

import pandas as pd

from tabarena.benchmark.experiment import AGModelBagExperiment, ExperimentBatchRunner