增量训练出来的分词模型分词输出是二维的

In [7]: my_lac.run('福建厦门')
Out[7]: [['福建', '厦门'], ['', '']]

In [8]: lac.run('福建厦门')
Out[8]: ['福建', '厦门']

The report only provides Python examples comparing my_lac.run('福建厦门') with lac.run('福建厦门'); it names no source file, test, or entry point. Start by locating the incremental-training implementation and the run method, then reproduce the shown output and verify that incremental training returns the same one-dimensional segmentation format.

Written by the indexing model from the issue text.