MCPcopy Create free account
hub / github.com/DeepRec-AI/DeepRec / Feature

Method Feature

tensorflow/core/kernels/sparse_cross_op.cc:78–83  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

76// InternalType is int64 only when using HashCrosser.
77template <>
78int64 SparseTensorColumn<int64>::Feature(int64 batch, int64 n) const {
79 const int64 start = feature_start_indices_[batch];
80 if (DT_STRING == values_.dtype())
81 return Fingerprint64(values_.vec<tstring>().data()[start + n]);
82 return values_.vec<int64>().data()[start + n];
83}
84
85// InternalType is string or StringPiece when using StringCrosser.
86template <>

Callers 15

_int64_featureMethod · 0.45
_bytes_featureMethod · 0.45
_recordMethod · 0.45
_int64_featureFunction · 0.45
_bytes_featureFunction · 0.45
_int64_featureFunction · 0.45
_bytes_featureFunction · 0.45
test_parse_exampleMethod · 0.45
test_parse_exampleMethod · 0.45

Calls 4

Fingerprint64Function · 0.85
to_stringFunction · 0.85
dtypeMethod · 0.45
dataMethod · 0.45

Tested by 15

_int64_featureMethod · 0.36
_bytes_featureMethod · 0.36
_recordMethod · 0.36
_int64_featureFunction · 0.36
_bytes_featureFunction · 0.36
test_parse_exampleMethod · 0.36
test_parse_exampleMethod · 0.36
test_parse_exampleMethod · 0.36
test_parse_exampleMethod · 0.36