Error happen when using LightGBM with Pipeline provided by sklearn

[PaperStrange]

Describe the bug
Using LightGBM model as a classfier of pipeline cause a error, saying checking dimension failed as "ValueError: bad input shape" when call ".fit" of the pipeline

To Reproduce
Steps to reproduce the behavior:

1. Go to 'DS_term2' branch
2. Click on 'DS_term2_project2/models/train_classifier.jpynb'
3. Scroll down to 'evaluate_model' title
4. See error displayed in the second cell below the title cell

Expected behavior
Normal running procedure

Screenshots

Desktop (please complete the following information):

• OS: Ubuntu 14
• Browser chrome
• Version 43

Smartphone (please complete the following information):

• Device: [e.g. iPhone6]
• OS: [e.g. iOS8.1]
• Browser [e.g. stock browser, safari]
• Version [e.g. 22]

Additional context
None yet

[PaperStrange]

2019/3/27: refer to the stackflow, sklearn Pipeline with multi-class classification problem may not work proper with a multi-class classifier