/alert_multimodal_ml

Primary LanguageJupyter Notebook

ML_skyportal

  • kowalski.py: Loads the Kowalski dataset.

  • processor/:

    • photometry_processor.py: Preprocesses photometry data.
    • alert_processor.py: Handles alerts, including images and metadata.
    • data_processor.py: General data preprocessing tasks.

  • supernova_dataset.py: Aggregates functions from processors for comprehensive data preparation.

  • data_generator.py: Inherits from Keras, manages data during model training.

  • Data Loading and Preprocessing:

    • Use kowalski.py to load data.
    • Utilize processor classes for specific preprocessing needs.
    • supernova_dataset.py combines all preprocessing steps.

  • Data Generation for Training:

    • data_generator.py provides batch data generation for Keras training.

  • Detailed Function Usage:

    • Refer to the Show_data notebook for detailed instructions on each function, from loading data to training a model.