merlin.models.tf.TensorInitializer

class merlin.models.tf.TensorInitializer(weights: Union[tensorflow.python.framework.ops.Tensor, Any], **kwargs)

Bases: keras.initializers.initializers.Initializer

Initializer that returns a tensor (e.g. pre-trained embeddings) set in the constructor

__init__(weights: Union[tensorflow.python.framework.ops.Tensor, Any], **kwargs)

Methods

__init__(weights, **kwargs)
from_config(config)	Instantiates an initializer from a configuration dictionary.
from_dataset(data, **kwargs)
get_config()

classmethod from_dataset(data: Union[merlin.io.dataset.Dataset, pandas.core.frame.DataFrame], **kwargs) → merlin.models.tf.utils.tf_utils.TensorInitializer

get_config()