sequence_expand operator shouldn't tailor input's y-axis dimension
Created by: pkuyym
An example:
Input LoDTensor:
shape: [2,5,]
LoD: [[ 0,1,2, ]]
data: 0.1,0.3,0.2,0.15,0.25,0.2,0.15,0.25,0.1,0.3
Referred LoDTensor:
shape: [2,1,]
LoD: [[ 0,1,2, ][ 0,1,2, ]]
data: 1,1
Output LoDTensor:
shape: [2,1,]
LoD: [[ 0,1,2, ][ 0,1,2, ]]
data: 0.1,0.3However, the expected output should be:
Output LoDTensor:
shape: [2,5]
LoD: [[ 0,1,2, ][ 0,1,2, ]]
data: 0.1,0.3,0.2,0.15,0.25,0.2,0.15,0.25,0.1,0.3