Python 3.6+
installedscikit-learn
and joblib
installedrun
in Neptune to which you can log metadata.project=my_workspace/my_project
: your workspace name and project name,api_token=YOUR_API_TOKEN
: your Neptune API token.api_token
argument.PARAMS
dictionary with all the parameters that you want to keep track of."train/loss"
is a name of the log with a hierarchical structure."train/loss"
is a series of values - you can log multiple values to this log."train/acc"
, "val/f1_score"
, "train/log-loss"
, "test/acc"
).log()
method is the actual value you want to log.run["train/loss"].log(loss)
multiple times at each step."test/acc"
..track_files()
method. Just pass the path to the file you want to log to Neptune.stop()
method. This is needed only while logging from a notebook environment. While logging through a script, Neptune automatically stops tracking once the script has completed execution.