Summary: 13 instances, 12 unique

Text	Count
# TODO: limit the size of memo	1
# TODO: #language_specific	1
# TODO: Move this to setup or add the folders to the git repo	1
# TODO: Penalize less deletion of unimportant words	2
return np.log(1 + len(get_word2rank()))  # TODO: This is completely arbitrary	1
# TODO: Is having a vector of zeros the best embedding for unknown words?	1
# TODO: Duplicate of get_addition_proportion, same for deletion	1
# TODO: Default value is arbitrary	1
# TODO: Handle this case	1
# TODO: Can't reproduce the task's formula for RMSE	1
# TODO: This class might be better of somewhere else	1
# TODO: Fix LM features	1