solutions

Nice work!
The version of implementation can reach 22+ BLUE score. However, my implementation have only 0.16+ BLUE score on test dataset. Comparing with your work, I found changing the concatenation torch.cat((Y_t, o_pre), dim=1) to torch.cat((o_pre, Y_t), dim=1) can only reach 0.16+ BLUE score.

B.W.T. chaging the concatenating order between dec_hidden and a_t in step function also

The description md files can be copy paste from browser.

Example:

Before:

-- 7 Billion Humans (2145) --
-- 46: Compulsory Office Romance --

a:
if e == worker
  tell e Hi!
  listen Hi!
else
  listen Hi!
  tell w Hi!
jump a

After:

-- 7 Billion Humans (2231) --
-- 46: Compulsory Office Romance --

a:
if e == worker:
	tell e hi
	listenfor hi
else:
	listenfor hi
	tell w hi
endif
jump a

I will try to find