simple_feature_store.py

from typing import Iterable, Callable, Dict

class SuperSimpleFeatureStore:
    
    def __init__(self, dataframe: pd.DataFrame):
        self.dataframe = dataframe
        self.feature_dict = dataframe.set_index("user_id").T.to_dict()
            
    def register_feature(self, name: str, feature_definition: Callable) -> None:
        for key in self.feature_dict:
            self.feature_dict[key][name] = feature_definition(self.feature_dict[key]) 
            
    def get_user_feature(self, user_id: str) -> Dict:
        return self.feature_dict.get(user_id)
    
    def get_all_data(self) -> pd.DataFrame:
        return pd.DataFrame(self.feature_dict).T

Hey @kylegallatin I came across your post on Medium today! Great tutorial, thanks a lot for clearing up how ML works for production!

That said, I think there might be an error in line 7. I actually don't see any user_data variable declared or initialized before this occurence. 😅

@anuran-roy thanks for bringing it up! The gists were copied directly from the notebook here that's meant to be runnable e2e - but you're right that as a standalone gist keeping those things as global variables isn't very useful. I'll update it!

Oh but that is actually just a straight up error, using the constructor argument on one line and a previously defined variable on another. Thanks for bringing it up! I've fixed it.