Where does `df.to_csv(...)` go?


#1

When I run a job with neptune send... and use a pandas df.to_csv(...) statement to write out a file, without specifying a path, where does that file go? How do I access that file?


#2

According to a pandas documentation, in this case the result is returned as a string.
See: https://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.to_csv.html


#3

Sorry, I should have been more clear:

On Neptune if I say df.to_csv('outfile.csv') where in the Neptune path structure does that file go?


#4

Hi,

If you’re running locally by neptune run, your file would go to your current directory.

If you’re running remotely by neptune send or via notebook, your files end up in /neptune/ directory.

Keep in mind, changes in /neptune directory are not saved after experiment ends!

It’s best to copy all important files to /output so we can store them for you :slight_smile: