Dragon API Dragon v0.3.0 Python GlorotNormal¶ class dragon.vm.tensorflow.keras.initializers.GlorotNormal(dtype='float32')[source]¶ Fill tensor from a glorot normal distribution. \[\text{tensor} \sim \mathcal{N}(0, \frac{2}{\text{fan\_avg}}) \] __init__¶ GlorotNormal.__init__(dtype='float32')[source]¶ Create a GlorotNormal initializer. Parameters: dtype (str, optional, default='float32') – The data type to set as default. Methods¶ __call__¶ VarianceScaling.__call__( shape, dtype=None, **kwargs)[source] Return a tensor initialized from the initializer. Parameters: shape (Sequence[int]) – The tensor shape. dtype (str, optional) – The optional data type. Returns:dragon.Tensor – The output tensor.