I'm using 3 datasets to create EntitySet using featuretools and use deep feature synthesis to generate additional features:
entity_set = ft.EntitySet("basketball_players")
entity_set.add_dataframe(dataframe_name="player_data",
dataframe=player_data,
index='name'
)
entity_set.add_dataframe(dataframe_name="players",
dataframe=players,
index='name'
)
entity_set.add_dataframe(dataframe_name="season_stats",
dataframe=season_stats,
index='name'
)
feature_matrix, features = ft.dfs(target_dataframe_name="season_stats",
entityset=entity_set,
max_depth = 3
)
However, when I do feature_matrix.columns, I don't see that any new features were generated. I checked the documentation, and it is not necessary to add explicitly add trans_primitives and agg_primitives: there are
trans_primitives (list[str or TransformPrimitive], optional) –
List of Transform Feature functions to apply.
Default: [“day”, “year”, “month”, “weekday”, “haversine”, “num_words”, “num_characters”]
and
agg_primitives (list[str or AggregationPrimitive], optional) –
List of Aggregation Feature types to apply.
Default: [“sum”, “std”, “max”, “skew”, “min”, “mean”, “count”, “percent_true”, “num_unique”, “mode”]
So new features should be generated anyway. What I'm doing wrong?