Training error

[Haithienbao]

When i try to train this code, i get error like this:

data created
data loaded
data created
data loaded
init trainer
training
=> Training epoch # 1
/torch/install/bin/luajit: /torch/install/share/lua/5.1/threads/threads.lua:183: [thread 2 callback] ./datasets/lane.lua:22: attempt to index local 'label' (a number value)
stack traceback:
./datasets/lane.lua:22: in function 'get'
./dataloader.lua:96: in function <./dataloader.lua:92>
[C]: in function 'xpcall'
/torch/install/share/lua/5.1/threads/threads.lua:234: in function 'callback'
/torch/install/share/lua/5.1/threads/queue.lua:65: in function </torch/install/share/lua/5.1/threads/queue.lua:41>
[C]: in function 'pcall'
/torch/install/share/lua/5.1/threads/queue.lua:40: in function 'dojob'
[string " local Queue = require 'threads.queue'..."]:15: in main chunk
stack traceback:
[C]: in function 'error'
/torch/install/share/lua/5.1/threads/threads.lua:183: in function 'dojob'
./dataloader.lua:157: in function '(for generator)'
./train.lua:58: in function 'train'
main.lua:51: in main chunk
[C]: in function 'dofile'
/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk
[C]: at 0x55c09631d570
load image failed!
load image failed!

I run this code in google colab, cuda10,
Could you have me solve this problem?
Thank you!