Design a Robust Cross-Validation Strategy for Your Machine Learning Project

You are an expert machine learning engineer. Help me design a comprehensive cross-validation strategy for my project with the following details: **Dataset Description:** - Dataset name/domain: [DATASET_NAME_OR_DOMAIN] - Number of samples: [NUM_SAMPLES] - Number of features: [NUM_FEATURES] - Target variable type: [CLASSIFICATION/REGRESSION] - Is the data time-series or sequential? [YES/NO] - Is the dataset imbalanced? [YES/NO, and approximate class ratios if applicable] **Project Goal:** [DESCRIBE_YOUR_PREDICTION_GOAL] Please provide the following in your response: 1. **Recommended CV Method:** Choose the most appropriate cross-validation technique (e.g., k-fold, stratified k-fold, time-series split, leave-one-out, group k-fold, nested CV) and justify why it suits my data. 2. **Number of Folds/Splits:** Recommend a specific number of folds with reasoning based on my dataset size and computational constraints. 3. **Data Leakage Prevention:** Identify potential sources of data leakage specific to my use case and explain how the CV design mitigates them. 4. **Evaluation Metrics:** Suggest 2-3 evaluation metrics to track across folds, explaining why each is appropriate. 5. **Implementation Outline:** Provide a Python code skeleton using scikit-learn that implements the recommended CV strategy, including proper preprocessing within folds using pipelines. 6. **Interpreting Results:** Explain how to interpret the mean and standard deviation of scores across folds, and what thresholds would indicate overfitting or high variance. 7. **Common Pitfalls:** List 3 common mistakes people make with cross-validation in similar projects and how to avoid them.