Spaces:
Running
Running
File size: 2,381 Bytes
211c37c | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 | #!/usr/bin/env python3
"""
main.py - CLI entry point for the Tabular AutoML framework.
Usage:
python main.py --data dataset.csv --target churn --task classification
python main.py --data housing.csv --target price --task regression --time_budget 120
"""
import argparse
import sys
import os
import pandas as pd
# Allow running from root directory
sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
from automl import AutoML
def parse_args():
parser = argparse.ArgumentParser(
description="Tabular AutoML Framework",
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
parser.add_argument("--data", required=True, help="Path to CSV dataset")
parser.add_argument("--target", required=True, help="Target column name")
parser.add_argument(
"--task",
required=True,
choices=["classification", "regression"],
help="Task type",
)
parser.add_argument(
"--time_budget",
type=float,
default=None,
help="Training time budget in seconds (optional)",
)
parser.add_argument(
"--n_trials",
type=int,
default=15,
help="Number of Optuna HPO trials per model",
)
parser.add_argument(
"--output_dir",
default="./automl_output",
help="Directory to save outputs",
)
parser.add_argument(
"--no_feature_engineering",
action="store_true",
help="Disable feature engineering",
)
parser.add_argument("--seed", type=int, default=42)
return parser.parse_args()
def main():
args = parse_args()
# Load data
print(f"\nLoading dataset: {args.data}")
try:
df = pd.read_csv(args.data)
except Exception as e:
print(f"Error loading CSV: {e}")
sys.exit(1)
if args.target not in df.columns:
print(f"Error: target column '{args.target}' not found.")
print(f"Available columns: {list(df.columns)}")
sys.exit(1)
# Run AutoML
automl = AutoML(
task_type=args.task,
time_budget=args.time_budget,
n_optuna_trials=args.n_trials,
output_dir=args.output_dir,
seed=args.seed,
use_feature_engineering=not args.no_feature_engineering,
)
automl.fit(df, target_col=args.target)
automl.report()
if __name__ == "__main__":
main()
|