Summary: 24 instances, 24 unique

Text	Count
# TODO make option to go from tokenized version, not necessary to retokenized the whole thing, a fonction "Select starting point"	1
# TODO add restore BPE of XLM utils into that class	1
# TODO make this cleaner with tree sitter AST ?	1
pred_mask[0] = 0  # TODO: remove	1
# TODO check number of lines	1
a StopIteration error (TODO: Make it return None)	1
# TODO: Annotations (for everything x:)	1
# TODO check that sentences are < max_len like in deobfuscate_by_variable	1
# TODO: check type for variable definitions?	1
# TODO: Annotations (for everything x:)	1
# TODO: I have no idea if this functions in not Python 3.5	1
# TODO raise ValueError(f'Invalid python function \n {code}\n')	1
# TODO: Name should eventually be unique per scope, as we	1
):  # AMP checkpoint reloading is buggy, we cannot do that - TODO: fix - https://github.com/NVIDIA/apex/issues/250	1
TODO: If we need more granularity, we need to edit how this works (would need	1
TODO: Properties that are not defined (that are None) just come back as blanks,	1
# TODO:keyword (in Python <3.3)	1
# TODO: After coming back to this a second time, the names aren't really sticking,	1
# TODO: This could be better an iterator, not a list return	1
# TODO could make it O(log(n)) with binary search for find first token	1
TODO: Convert to iterator, not list generator	1
# TODO: only output when necessary - len(params.clm_steps + params.mlm_steps) > 0	1
# TODO -> do it cleaner way for opensourcing	1
# TODO clean lang1	1