megnet.config module¶
Data types
- class DataType[source]¶
Bases:
object
Data types for tensorflow. This enables users to choose from 32-bit float and int, and 16-bit float and int
- np_float¶
alias of
numpy.float32
- np_int¶
alias of
numpy.int32
- classmethod set_dtype(data_type: str) None [source]¶
Class method to set the data types :param data_type: ‘16’ or ‘32’ :type data_type: str
- tf_float = tf.float32¶
- tf_int = tf.int32¶