Description

nquinn408

opened on Apr 21, 2026

Issue body actions

Is your feature request related to a problem? Please describe.
Currently, feast supports dynamic aggregation through ODFV from raw data source. For variable window aggregation, the features are independently computed so there is no raw source but aggregation is still being read / written.

Describe the solution you'd like
We want to allow users to not define raw data source but instead replace data source with new optional input_schema field to define raw data input type. For example, instead of passing in txn_amount from external feature view, it is submitted as input_schema. See example below:

@on_demand_feature_view(
      input_schema=[
          Field(name="txn_amount", dtype=Float64),
      ],
      schema=[
          Field(name="txn_count", dtype=Int64),
          Field(name="total_txn_amount", dtype=Float64),
          Field(name="avg_txn_amount", dtype=Float64),
      ],
      aggregations=[
          Aggregation(column="txn_amount", function="count", name="txn_count",
                      window=timedelta(days=30)),
          Aggregation(column="txn_amount", function="sum", name="total_txn_amount",
                      window=timedelta(days=30)),
          Aggregation(column="txn_amount", function="mean", name="avg_txn_amount",
                      window=timedelta(days=30)),
      ],
      entities=[user],
      write_to_online_store=True,
  )

This allows dynamic systems like cassandra or iceberg that support time series data representations, to support variable window engines which support writing and reading to / from same source.

Describe alternatives you've considered
Using dummy source (confusing, needlessly complex), using aggregation on feature view (not supported)

Activity

1. nquinn408

   added

   kind/featureNew feature or requestNew feature or request

   on Apr 21, 2026
2. franciscojavierarceo commented on Apr 21, 2026

   

   franciscojavierarceo

   on Apr 21, 2026

   Member

   More actions

   @HaoXuAI @ntkathole want your take here
3. nquinn408

   added 3 commits that reference this issue on Apr 21, 2026

   feat: Optional input_schema for ODFV (feast-dev#6308)

   47a8d31

   feat: Optional input_schema for ODFV (feast-dev#6308)

   56b0620

   feat: Optional input_schema for ODFV (#6308) (#6312)

   f08b4e8
4. ntkathole commented on Apr 25, 2026

   

   ntkathole

   on Apr 25, 2026

   Member

   More actions

   @nquinn408 Looks good to me, Also since this is a pure aggregation view with no udf, it can be defined via direct instantiation instead of decorator, that would avoid serialization overhead of empty function

   txn_stats = OnDemandFeatureView(
       name="compute_txn_stats",
       input_schema=[Field(name="txn_amount", dtype=Float64)],
       schema=[
           Field(name="txn_count",        dtype=Int64),
           Field(name="total_txn_amount", dtype=Float64),
           Field(name="avg_txn_amount",   dtype=Float64),
       ],
       aggregations=[
           Aggregation(column="txn_amount", function="count",
                       name="txn_count",        time_window=timedelta(days=30)),
           Aggregation(column="txn_amount", function="sum",
                       name="total_txn_amount", time_window=timedelta(days=30)),
           Aggregation(column="txn_amount", function="mean",
                       name="avg_txn_amount",   time_window=timedelta(days=30)),
       ],
       entities=[user],
       write_to_online_store=True,
   )
   
5. franciscojavierarceo

   added a commit that references this issue on May 4, 2026

   chore(release): release 0.63.0

   50ad181
6. singhhimanshu0811 commented on Jun 30, 2026

   

   singhhimanshu0811

   on Jun 30, 2026

   Contributor

   More actions

   hey @franciscojavierarceo is this still open? if yeah will take this up. pls let me know!!

   Reacted by Francisco Javier Arceo

Issue actions