Cookbook

Personal reference scripts for commonly used code

Cookbook

Personal reference scripts for commonly used code

• Machine Learning: A folder containing scripts for commonly used machine learning code
  • Preprocessing.py: Preparing data for machine learning tasks, primarily using pandas and sklearn
  • scikit-learn: Also includes LightGBM and XGBoost
    • ModelTraining.py: Cross validation, hyperparameter tuning, feature selection, etc.
    • Evaluation.py: Evaluation plots, collecting eval metrics, learning curves, feature importance, etc.
    • LighTGBM.py: Early stopping and other code that’s convenient to copy/paste
  • TensorFlow
    • Keras.py: Commonly used code for Keras
    • KerasMNIST.py: Training a convolutional net on the MNIST data with Keras
    • TensorFlowMNIST.py: Training a convolutional net on the MNIST data with TensorFlow
  • PyTorch
    • PyTorch.py: Commonly used code for PyTorch
    • PyTorchMNIST.py: Training a convolutional net on the MNIST data with PyTorch
  • SparkML
    • SparkML.py: Commonly used code for SparkML. Includes preprocessing, hyperparameter tuning, cross validation, and so on.
• Plotting: Code snippets for common plots
  • Matplotlib.py
  • Plotly.py
• Misc: For scripts that don’t fit within any other folders
  • EDA.py: EDA reports, missing values, and outliers
  • NLP.py: NLTK natural language processing tasks
  • PySpark.py: Missing values, datatype conversions, encoding categorical columns, and prepping data for models
• DevOps: A folder containing scripts for operationalizing machine learning models
  • Flask: Operationalizing a trained machine learning model as a RESTful API
    • Web App
    • Sending Requests