$OpDocParsingDecodeCSV
$OpDocParsingDecodeCSV
STRING tensor where each string is a record/row in the csv and all records should have the same format.
One tensor per column of the input record, with either a scalar default value for that column or empty if the column is required.
Output tensor data types.
Delimiter used to separate fields in a record.
If false
, the op treats double quotation marks as regular characters inside the
string fields (ignoring RFC 4180, Section 2, Bullet 5).
Name for the created op.
Created op outputs.
IllegalArgumentException
If records
is not a STRING tensor.
$OpDocParsingDecodeJSONExample
$OpDocParsingDecodeJSONExample
STRING tensor where each string is a JSON object serialized according to the JSON mapping
of the Example
proto.
Name for the created op.
Created op output.
IllegalArgumentException
If jsonExamples
is not a STRING tensor.
$OpDocParsingDecodeRaw
$OpDocParsingDecodeRaw
STRING tensor interpreted as raw bytes. All the elements must have the same length.
Output tensor data type.
Boolean value indicating whether the input bytes
are stored in little-endian order. Ignored
for dataType
values that are stored in a single byte, like UINT8.
Name for the created op.
Created op output.
IllegalArgumentException
If bytes
is not a STRING tensor.
$OpDocParsingDecodeTensor
$OpDocParsingDecodeTensor
STRING tensor containing a serialized TensorProto
proto.
Data type of the serialized tensor. The provided data type must match the data type of the serialized tensor and no implicit conversion will take place.
Name for the created op.
Created op output.
IllegalArgumentException
If data
is not a STRING tensor.
$OpDocParsingEncodeTensor
$OpDocParsingEncodeTensor
Tensor to encode.
Name for the created op.
Created op output.
$OpDocParsingStringToNumber
Contains functions for constructing ops related to parsing data.