Doubt in the coding part of Encoder Decoder architecture

In the inference part of the model (before training), I have a confusion with this line of code.

out = infer(net, 'INDIA', 30)

What is the 30 in the code represent? Whether “INDIA” is encoded and mapped to 30 characters of Hindi in the decoder part? But is it not like if “INDIA” is the input which is encoded then the output will possibly be having 6 characters as the output in Hindi?

Please help to clarify in this regard.

The infer() function was missing in the notebook I guess.
Please check if this helps:

30 in your case represents the maximum number of output characters.

Just to clarify this means 30 is the maximum number of nodes in the output layer of the network, where if there is an output which is less than 30 characters (say 25 characters) then the remaining 5 characters will be filled by <pad> when the model is well trained?