How to get the 114 features of the training dataset

[manwu1994]

Hello, thank you for your work first.
I would like to how to get the 114 features of the training dataset.
And what does the "# index 0, 4, 5: duration, src_bytes, dst_bytes (in kyoto: index 0, 2, 3)" mean in your comments?

Thank you so much.