TensorFlow 1 version View source on GitHub

Representation of HDF5 dataset to be used instead of a Numpy array.

    datapath, dataset, start=0, end=None, normalizer=None

THIS CLASS IS DEPRECATED. Training with HDF5Matrix may not be optimized for performance, and might not work with every distribution strategy.

We recommend using to load your HDF5 data into a Dataset and passing that dataset to Keras.


  • datapath: string, path to a HDF5 file
  • dataset: string, name of the HDF5 dataset in the file specified in datapath
  • start: int, start of desired slice of the specified dataset
  • end: int, end of desired slice of the specified dataset
  • normalizer: function to be called on data when retrieved


  • dtype: Gets the datatype of the dataset.

  • ndim: Gets the number of dimensions (rank) of the dataset.

  • shape: Gets a numpy-style shape tuple giving the dataset dimensions.

  • size: Gets the total dataset size (number of elements).


ImportError if HDF5 & h5py are not installed



View source



View source


Class Variables

  • refs