Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

How can I use Feast with only InputRequest source #3660

Open
htluandc2 opened this issue Jun 23, 2023 · 0 comments
Open

How can I use Feast with only InputRequest source #3660

htluandc2 opened this issue Jun 23, 2023 · 0 comments
Labels
kind/discussion kind/question Further information is requested

Comments

@htluandc2
Copy link

htluandc2 commented Jun 23, 2023

Is your feature request related to a problem? Please describe.
I can't use only RequestSource when make new features.

Describe the solution you'd like
Hi there,
I'm on a competition programming. I want to make some new features for my model. But I can't only using RequestSource on FeatureService. I must have least one FeatureView for FeatureService.

Describe alternatives you've considered
In short term, I make an FeatureView with one dummy column, like it:

fv = FeatureView(
    name='fv',
    entities=[entity],
    source=source,
    schema=[
         # I maked an dummy column 
        Field(name="dummy", dtype=feast.types.Float32),
    ]
)

features_v1 = FeatureService(
    name="features_v1",
    features=[
        fv,
        make_num_of_feature1,
    ]
)

But this solution is very slow in my competition.

Additional context
My code:

features_v1 = FeatureService(
    name="features_v1",
    features=[
        # fv,
        make_num_of_feature1,
    ]
)

My issue when I applied only InputRequest on FeatureSerivce:

f, validation_reference, timeout)
     64 def to_df(
     65     self,
     66     validation_reference: Optional["ValidationReference"] = None,
     67     timeout: Optional[int] = None,
     68 ) -> pd.DataFrame:
     69     """
     70     Synchronously executes the underlying query and returns the result as a pandas dataframe.
     71 
   (...)
     77         timeout (optional): The query timeout if applicable.
     78     """
---> 79     features_df = self._to_df_internal(timeout=timeout)
     81     if self.on_demand_feature_views:
     82         # TODO(adchia): Fix requirement to specify dependent feature views in feature_refs
...
   5900 ):
   5901     return self[name]
-> 5902 return object.__getattribute__(self, name)

AttributeError: 'DataFrame' object has no attribute 'persist'
@htluandc2 htluandc2 added the kind/feature New feature or request label Jun 23, 2023
@jeremyary jeremyary added kind/question Further information is requested kind/discussion and removed kind/feature New feature or request labels Mar 4, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
kind/discussion kind/question Further information is requested
Projects
None yet
Development

No branches or pull requests

2 participants