Categorize each feature by its primitive type in a set of primitives along with any dependencies

Checks if any logical types or semantic tags occur in a list of Woodwork input types and keeps track of unique input types.

If a feature is a DirectFeature, finds the primitive of the "original" base feature.

Calculates all LiteFeature combinations using the given hashmap of existing features, and the input set of required columns. Args: f

For a given primitive, find all feature sets that can be used to create new feature Args: feature_collection (FeatureCollection):

Helper function that selects all primitives that are instances of `primitive_kind`

Provides metrics for a list of primitives.

Filter a dataframe for all instances before time_last. If the dataframe does not have a time index, return the original dataf

Is there any other relationship with same parent and child dataframes?

Sets the secondary time index for a Woodwork dataframe passed in

If the saved schema version is newer than the current featuretools schema version, this function will output a warning saying so. If the

A list of pairs of the edges from this node and the nodes they point to. Examples: >>> from featuretools.utils i

Generate a hashing key from a Columns Schema. For example: - ColumnSchema(logical_type=Double) -> "Double" - ColumnSchema(semantic_tags={

Construct a primitive from the given dictionary (output from serialize_primitive).

Generate hashing keys from LiteFeature. For example: - LiteFeature("f1", Double, {"numeric"}) -> ['Double', 'numeric', 'Double,numeri

Finds whether time best corresponds to a value in days, hours, minutes, or seconds.

Flattens nested column schema inputs into a single list.

Returns system information as a dict

Returns a DataFrame that lists and describes each built-in primitive.

Load the Australian daily-min-temperatures weather dataset. Args: nrows (int): Passed to nrows in ``pd.read_csv``. return_s

Read data description from disk, S3 path, or URL. Args: path (str): Location on disk, S3 path, or URL to read `data_description.json`.

Select features that have at least 2 unique values and that are not all null Args: feature_matrix (:class:`pd.DataFrame`): DataFrame whos

Add other

Divide by other

Multiply by other

Merge ancestor_relationship_columns from parent_df into child_df, adding a prefix to each column name specifying the relationship.

Add a column to the cutoff df linking it to the dataframe at the end of the path. Return the updated cutoff df and the name of this colu

Add the given feature to the root of the trie, and recurse on its dependencies. If it is in approximate_feature_trie then it will not

converts all columns from the given dataframe into features Args: all_features (dict[dataframe name -> dict[str -> BaseFeature]])

Build the feature trie by adding the target features and their dependencies recursively.

Iterates over the input dictionary to build the ignore_columns defaultdict. Expects the input_dict's keys to be strings, and values to be lists of

Traverses all identity features and creates a Compare for each one, based on some heuristics Args: all_features (dict

Generate dataframes with features calculated for this node of the trie, and all descendant nodes. The dataframes will be stored in df

Check that the cutoff time values are of the proper type given the entityset time type