comparison pycaret_train.py @ 10:49f73a3c12f3 draft default tip

planemo upload for repository https://github.com/goeckslab/gleam commit 1ffd143e57fa952ee9dd84fc141771520aea0791

9:e7dd78077b72

import argparse
import logging

from pycaret_classification import ClassificationModelTrainer
from pycaret_regression import RegressionModelTrainer

logging.basicConfig(level=logging.DEBUG)

        type=int,
        default=42,
        help="Random seed for PyCaret setup",
    )
    parser.add_argument(
        "--probability_threshold",
        type=float,
        default=None,
        help="Probability threshold for classification decision,",
    )

        default=None,
        help="Metric used to select the best model (e.g. AUC, Accuracy, R2, RMSE).",
    )

    args = parser.parse_args()

    # Normalize cross-validation flags: --no_cross_validation overrides --cross_validation
    if args.no_cross_validation:
        args.cross_validation = False
    # If --cross_validation was passed, args.cross_validation is True

        "polynomial_features": args.polynomial_features,
        "feature_interaction": args.feature_interaction,
        "feature_ratio": args.feature_ratio,
        "fix_imbalance": args.fix_imbalance,
        "tune_model": args.tune_model,
        "probability_threshold": args.probability_threshold,
        "best_model_metric": args.best_model_metric,
    }
    LOG.info(f"Model kwargs: {model_kwargs}")

10:49f73a3c12f3

import argparse
import logging
import os

from pycaret_classification import ClassificationModelTrainer
from pycaret_regression import RegressionModelTrainer

logging.basicConfig(level=logging.DEBUG)

        type=int,
        default=42,
        help="Random seed for PyCaret setup",
    )
    parser.add_argument(
        "--n-jobs",
        dest="n_jobs",
        type=int,
        default=None,
        help="Number of parallel jobs; defaults to GALAXY_SLOTS or 1 if unset/invalid.",
    )
    parser.add_argument(
        "--probability_threshold",
        type=float,
        default=None,
        help="Probability threshold for classification decision,",
    )

        default=None,
        help="Metric used to select the best model (e.g. AUC, Accuracy, R2, RMSE).",
    )

    args = parser.parse_args()

    # Derive n_jobs from CLI or GALAXY_SLOTS env var
    if args.n_jobs is not None:
        n_jobs = args.n_jobs
    else:
        slots_str = os.environ.get("GALAXY_SLOTS")
        try:
            n_jobs = int(slots_str) if slots_str is not None else 1
        except ValueError:
            n_jobs = 1

    # Normalize cross-validation flags: --no_cross_validation overrides --cross_validation
    if args.no_cross_validation:
        args.cross_validation = False
    # If --cross_validation was passed, args.cross_validation is True

        "polynomial_features": args.polynomial_features,
        "feature_interaction": args.feature_interaction,
        "feature_ratio": args.feature_ratio,
        "fix_imbalance": args.fix_imbalance,
        "tune_model": args.tune_model,
        "n_jobs": n_jobs,
        "probability_threshold": args.probability_threshold,
        "best_model_metric": args.best_model_metric,
    }
    LOG.info(f"Model kwargs: {model_kwargs}")