aligned.retrieval_job.DerivedFeatureJob

class documentation

class DerivedFeatureJob(RetrievalJob, ModificationJob): (source)

View In Hierarchy

Undocumented

Async Method compute_derived_features_pandas Undocumented
Async Method compute_derived_features_polars Undocumented
Method drop_invalid Drops invalid row based on the defined features.
Method filter Filters based on a condition referencing either a feature, a feature name, or an polars expression to filter on.
Method inject_store Undocumented
Method remove_derived_features Undocumented
Async Method to_lazy_polars Undocumented
Async Method to_pandas Undocumented
Class Variable requests Undocumented
Class Variable store Undocumented
Instance Variable job Undocumented
Property request_result Undocumented
Property retrieval_requests Undocumented

Inherited from RetrievalJob:

Static Method from_convertable Undocumented
Static Method from_dict Undocumented
Static Method from_lazy_function Undocumented
Static Method from_polars_df Undocumented
Method aggregate Undocumented
Method cache_raw_data Undocumented
Method cached_at Undocumented
Method chunked Undocumented
Method combined_features Undocumented
Method derive_features Undocumented
Method describe Undocumented
Method drop_duplicate_entities Undocumented
Method ensure_types Undocumented
Method fill_missing_columns Undocumented
Method ignore_event_timestamp Undocumented
Method join Undocumented
Method join_asof Undocumented
Method limit Undocumented
Method listen_to_events Undocumented
Method log_each_job Undocumented
Method monitor_time_used Undocumented
Method on_error SomeView.query().all().on_error(ValueError, lambda e: pl.DataFrame(...))
Method on_load Undocumented
Method polars_method Undocumented
Method rename Undocumented
Method return_invalid Undocumented
Method select Undocumented
Method select_columns Undocumented
Method split Undocumented
Async Method to_polars Undocumented
Async Method to_spark Undocumented
Method train_test Undocumented
Method train_test_validate Undocumented
Method transform_polars Undocumented
Method unique_entities Undocumented
Method unique_on Undocumented
Method unpack_embeddings Undocumented
Method update_vector_index Undocumented
Method validate_entites Undocumented
Method with_request Undocumented
Method with_subfeatures Undocumented
Method without_derived_features Undocumented
Async Method write_to_source Writes the output of the retrieval job to the passed source.
Property loaded_columns Undocumented

Inherited from ModificationJob (via RetrievalJob):

Method copy_with Undocumented
async def compute_derived_features_pandas(self, df: pd.DataFrame) -> pd.DataFrame: (source)

Undocumented

async def compute_derived_features_polars(self, df: pl.LazyFrame) -> pl.LazyFrame: (source)

Undocumented

def drop_invalid(self, validator: Validator | None = None) -> RetrievalJob: (source)
overrides aligned.retrieval_job.RetrievalJob.drop_invalid

Drops invalid row based on the defined features.

```python @feature_view(...) class WhiteWine:

wine_id = UInt64().as_entity()

quality = Int32().lower_bound(1).upper_bound(10)

valid_wines = WhiteWine.drop_invalid({
"wine_id": [0, 1, 2, 3, 4], "quality": [None, 4, 8, 20, -10]

})

print(valid_wines) >>> {

"wine_id": [1, 2], "quality": [4, 8]

}

Args:
validator (Validator): A validator that can validate the data.
The default uses the PolarsValidator
Returns:
RetrievalJob: A new retrieval job with only valid rows.
def filter(self, condition: FilterRepresentable) -> RetrievalJob: (source)
overrides aligned.retrieval_job.RetrievalJob.filter

Filters based on a condition referencing either a feature, a feature name, or an polars expression to filter on.

def inject_store(self, store: ContractStore) -> RetrievalJob: (source)
overrides aligned.retrieval_job.RetrievalJob.inject_store

Undocumented

def remove_derived_features(self) -> RetrievalJob: (source)
overrides aligned.retrieval_job.RetrievalJob.remove_derived_features

Undocumented

async def to_lazy_polars(self) -> pl.LazyFrame: (source)
overrides aligned.retrieval_job.RetrievalJob.to_lazy_polars

Undocumented

async def to_pandas(self) -> pd.DataFrame: (source)
overrides aligned.retrieval_job.RetrievalJob.to_pandas

Undocumented

requests: list[RetrievalRequest] = (source)

Undocumented

store: ContractStore | None = (source)

Undocumented

job: RetrievalJob = (source)
overrides aligned.retrieval_job.ModificationJob.job

Undocumented

request_result: RequestResult = (source)
overrides aligned.retrieval_job.RetrievalJob.request_result

Undocumented

retrieval_requests: list[RetrievalRequest] = (source)
overrides aligned.retrieval_job.RetrievalJob.retrieval_requests

Undocumented