joblib-parallel-gridsearch

name	joblib-parallel-gridsearch
description	Perform parallel grid search using joblib. Use this skill whenever a task requires optimizing multiple hyperparameters simultaneously across CPU cores.

Parallel Grid Search with Joblib

Joblib is the standard library for multiprocessing in Python data science workloads.

Basic Usage

from joblib import Parallel, delayed
import itertools

def evaluate_params(param1, param2):
    # Perform expensive computation
    score = param1 + param2
    return {'p1': param1, 'p2': param2, 'score': score}

# Define grid
param_grid = list(itertools.product([1, 2, 3], [0.1, 0.2]))

# Run in parallel
results = Parallel(n_jobs=-1, verbose=10)(
    delayed(evaluate_params)(p1, p2) for p1, p2 in param_grid
)

Tips

• n_jobs=-1 uses all available CPU cores.
• Pre-load data in the main process before spawning parallel workers to save memory.
• Pass required shared data as arguments to your delayed function.