csvfiles. Follow the same data requirements as described in the Jupyter notebook guide.
referenceis the path to the reference data,
currentis the path to the current data,
outputis the path to the output folder,
configis the path to the configuration file,
pretty_printto print the JSON profile with indents (for profile only).
data_driftto estimate the data drift,
num_target_driftto estimate target drift for the numerical target
cat_target_driftto estimate target drift for the categorical target
regression_performanceto explore the performance of a regression model
classification_performanceto explore the performance of a classification model
prob_classification_performanceto explore the performance of a probabilistic classification model
config.jsonfile or a
config.yamlfile. This file configures the way of reading your input data and the type of the report.
csvfiles with headers and there is no
datecolumn in the data.
csvfiles with headers and
datetimecolumn. We also specified the
column_mappingdictionary, where we added information about the
none- no sampling will be applied
nth- each Nth row of the file will be taken. This option works together with the
nparameter (see the example with the Dashboard above)
random- random sampling will be applied. This option works together with
ratioparameter (see the example with the Profile above)