tf.Variable.SaveSliceInfo

Information on how to save this Variable as a slice.

tf.Variable.SaveSliceInfo(
    full_name=None,
    full_shape=None,
    var_offset=None,
    var_shape=None,
    save_slice_info_def=None,
    import_scope=None
)

Provides internal support for saving variables as slices of a larger variable. This API is not public and is subject to change.

Available properties:

  • full_name
  • full_shape
  • var_offset
  • var_shape

full_name Name of the full variable of which this Variable is a slice.
full_shape Shape of the full variable, as a list of int.
var_offset Offset of this Variable into the full variable, as a list of int.
var_shape Shape of this Variable, as a list of int.
save_slice_info_def SaveSliceInfoDef protocol buffer. If not None, recreates the SaveSliceInfo object its contents. save_slice_info_def and other arguments are mutually exclusive.
import_scope Optional string. Name scope to add. Only used when initializing from protocol buffer.

spec Computes the spec string used for saving.

Methods

from_spec

View source

@classmethod
from_spec(
    spec: str
) -> Self

Parses a SaveSliceInfo spec string and returns a SaveSliceInfo object.

Args
spec The tensor slice spec string according to the SaveSliceInfo.spec property. The spec contains the space-separated shape of the full variable, followed by colon-separated pairs of the variable's offset and shape, where each pair is comma-separated. For example, consider a variable whose full shape is [4 3 5], offset is [0 1 3], and shape is [4 1 2]. This variable's SaveSliceInfo.spec would be "4 3 5 0,4:1,1:3,2".

Returns
A SaveSliceInfo object containing the extracted information.

Raises
ValueError If the input string is not in the expected format.

to_proto

View source

to_proto(
    export_scope=None
)

Returns a SaveSliceInfoDef() proto.

Args
export_scope Optional string. Name scope to remove.

Returns
A SaveSliceInfoDef protocol buffer, or None if the Variable is not in the specified name scope.