Documentation
¶
Index ¶
Constants ¶
This section is empty.
Variables ¶
var ErrInvalidChecksum = errors.New("invalid crc")
ErrInvalidChecksum is returned if header or footer checksum is invalid
Functions ¶
This section is empty.
Types ¶
type Example ¶
An Example is a mostly-normalized data format for storing data for training and inference. It contains a key-value store (features); where each key (string) maps to a Feature message (which is oneof packed BytesList, FloatList, or Int64List).
func NewExampleFromMap ¶
NewExampleFromMap returns a new Example based on a given map of go values
type Feature ¶
Feature contains Lists which may hold zero or more values.
func NewFeature ¶
NewFeature returns a Feature from a given go value
type PredictionLogReader ¶
type PredictionLogReader struct {
// contains filtered or unexported fields
}
PredictionLogReader reads tensorflow_serving.PredictionLog from a given io.Reader
func NewPredictionLogReader ¶
func NewPredictionLogReader(r io.Reader) *PredictionLogReader
NewPredictionLogReader returns PredictionLogReader for a given io.Reader
func (*PredictionLogReader) Next ¶
func (plr *PredictionLogReader) Next() error
Next advances PredictionLogReader or return io.EOF
func (*PredictionLogReader) Unmarshal ¶
func (plr *PredictionLogReader) Unmarshal(out *tensorflow_serving.PredictionLog) error
Unmarshal unmarshals current PredictionLog from the reader
type TFRecordReader ¶
type TFRecordReader struct {
// contains filtered or unexported fields
}
TFRecordReader reads tf records from a given io.Reader
func NewTFRecordReader ¶
func NewTFRecordReader(reader io.Reader) *TFRecordReader
NewTFRecordReader returns a new TFRecordReader for a given io.Reader
func (*TFRecordReader) Bytes ¶
func (tfr *TFRecordReader) Bytes() []byte
Bytes returns the current bytes from the reader
func (*TFRecordReader) Next ¶
func (tfr *TFRecordReader) Next() error
Next returns the next TFRecord or io.EOF error if the files have been consumed
Source Files