How to know the argument type of torch built in library functions

[aMY]

image package has a function called colorize which can be invoked using image.float.colorize. What is the dimension of the arguments of the function colorize.
In my case,
image=require 'image'
-- input: greyscale image of type FloatTensor of size 1x50x50
colorimage=image.float.colorize(input)
The above gives the following error:


[string "colorimage=image.float.colorize(input)"]:1: bad argument #2 to 'colorize' (torch.FloatTensor expected, got userdata)
stack traceback:
    [C]: in function 'colorize'
    [string "colorimage=image.float.colorize(input)"]:1: in main chunk
    [C]: in function 'xpcall'
    /home/rajiv/torch/install/share/lua/5.1/trepl/init.lua:679: in function 'repl'
    ...ajiv/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:204: in main chunk
    [C]: at 0x00405e90