static_properties_from_df#
- static_properties_from_df(df, image_cols: Tuple = (), label_cols: Tuple = (), prediction_cols: Tuple = (), partial_image_cols: Tuple = ()) Dict[int, Dict[PropertiesInputType, Dict[str, Any]]][source]#
Transform the pre-calculated properties from a DataFrame to the expected dict format.
Read more about the excepted dict format for pre-calculated properties at https://docs.deepchecks.com/stable/user-guide/vision/vision_properties.html
- Parameters
- df: pd.Dataframe
A dataframe with the pre-calculated properties per sample, the indices should match those of the samples in the dataset.
- image_cols: Tuple, default: ()
The names of the columns containing the results of the properties calculated on images.
- label_cols: Tuple, default: ()
The names of the columns containing the results of the properties calculated on labels.
- prediction_cols: Tuple, default: ()
The names of the columns containing the results of the properties calculated on predictions.
- partial_image_cols: Tuple, default: ()
The names of the columns containing the results of the properties calculated on partial images - images cut out of the original images such as bounding boxes.
- Returns
- The static properties in the format expected by the checks.