data_measurements/dataset_statistics.py (7 lines):
	- line 47: # TODO: Read this in depending on chosen language / expand beyond english
	- line 128: # TODO: Not being used anymore; make sure & remove.
	- line 135: # TODO: Not being used anymore. Make sure and remove
	- line 305: # TODO: Tighten the rest of this similar to text_duplicates.
	- line 458: # TODO: Are we not using this anymore?
	- line 517: # TODO: Does z even need to be self?
	- line 575: # TODO: Add warnings (which words are missing) to log file?


data_measurements/npmi/npmi.py (4 lines):
	- line 34: # TODO: Should be possible for a user to specify this.
	- line 82: # TODO: Users ideally can type in whatever words they want.
	- line 89: # TODO: Let users specify
	- line 289: # TODO: Change this logic so just the vocabulary is given.


npmi/npmi.py (4 lines):
	- line 15: # TODO: Change print statements to logging?
	- line 47: # TODO: Is this necessary?
	- line 75: # TODO: Create docs for this.
	- line 186: # TODO: Is this better?


app.py (3 lines):
	- line 133: # TODO: If these are cached, can't we just show them by default?
	- line 165: # TODO: Fix how this is a weird outlier.
	- line 178: # # TODO: Make this less of a weird outlier.


data_measurements/labels/labels.py (3 lines):
	- line 20: # TODO: This should ideally be in what's returned from the evaluate library
	- line 164: # TODO: Handle the case where there are multiple label columns.
	- line 226: # TODO: Incorporate this summation into what the evaluate library returns.


utils/gradio_utils.py (3 lines):
	- line 124: # TODO @yacine: Explain what this is doing and why eg tp[0] could = "id"
	- line 280: # TODO: Check if this is slow when the size is large --
	- line 405: # TODO: Nice UI version of the content in the comments.


data_measurements/text_duplicates/text_duplicates.py (2 lines):
	- line 18: # This isn't in the evaluate measurement, but TODO to add that...
	- line 86: # TODO: Use df_to_html rather than write_json_as_html;


data_measurements/zipf/zipf.py (2 lines):
	- line 127: # TODO: These proportions may have already been calculated.
	- line 179: # TODO: This might fit better in its own file handling class?


run_data_measurements.py (2 lines):
	- line 58: # TODO: Catch error exceptions for each measurement, so that an error
	- line 304: # TODO: print out local or hub cache directory location.


widgets/text_lengths.py (1 line):
	- line 55: # TODO: Add text on choosing the length you want to the dropdown.


data_measurements/lengths/lengths.py (1 line):
	- line 100: # This is a hack to handle a UI display error (TODO: file bug)


data_measurements/embeddings/embeddings.py (1 line):
	- line 254: # TODO: batch across second dimension


data_measurements/perplexity/perplexity.py (1 line):
	- line 40: # TODO: What other stuff might be useful to grab?


utils/dataset_utils.py (1 line):
	- line 353: # TODO: Breaks the CLI if this isn't checked.