Feature store with integrated tracking
The present application discloses a method, system, and computer system for managing a plurality of features and storing lineage information pertaining to the features. The method includes obtaining one or more datasets, determining a first feature, wherein the first feature is determined based at least in part on the one or more datasets, and storing the first feature in a feature store. The first feature is stored in association with a dataset indication of the one or more datasets from which the first feature is determined. The feature store comprises a plurality of features.
1. A method comprising:
accessing source data stored in a data store, the source data comprising raw data;
generating a feature based on the source data;
training a set of machine learning models based on the feature;
storing the feature in a feature store;
storing, in the feature store, metadata in association with the feature, the metadata including: (i) a mapping from the feature to upstream lineage data indicating the source data used to generate the feature, and (ii) a mapping from the feature to downstream lineage data indicating the set of machine learning models trained based on the feature;
determining, based on the metadata in the feature store including the mapping from the feature to the upstream lineage data, that a modification has been made to the feature;
in response to determining that the modification has been made to the feature, identifying, based on metadata in the feature store including the mapping from the feature to the downstream lineage data: (i) a first machine learning model from the set of machine learning models trained based on the feature and (ii) a model serving endpoint that deploys the first machine learning model; and
transmitting a notification to the model serving endpoint that deploys the first machine learning model, the notification indicating that the feature has been modified, wherein the model serving endpoint is a web service.
2. The method of claim 1 , further comprising:
updating the mapping from the feature to the upstream lineage data based on the modification to the feature, wherein the mapping is updated to reflect modified upstream lineage data indicating revised source data.
3. The method of claim 1 , further comprising generating the mapping from the feature to the upstream lineage data indicating the source data used to generate the feature.
4. The method of claim 1 , further comprising:
determining that the feature is used to train a new machine learning model, the new machine learning model not in the set of machine learning models; and
updating the mapping from the feature to downstream lineage data indicating the new machine learning model.
5. The method of claim 1 , further comprising:
storing the first machine learning model in a model store along with the mapping from the feature to the downstream lineage data indicating that the feature was used to train the first machine learning model, the model store storing at least a second machine learning model different than the first machine learning model.
6. The method of claim 1 , wherein the mapping from the feature to the upstream lineage data further comprises code used to determine the feature.
7. The method of claim 1 , further comprising:
receiving data defining a second feature, the data defining the second feature identifying source data used to generate the second feature;
generating a mapping from the second feature to upstream lineage data indicating the source data used to generate the second feature; and
storing, in the feature store in association with the second feature, the mapping from the second feature to the upstream lineage data.
8. The method of claim 1 , further comprising:
sending, by the model serving endpoint: (i) a call to the feature store to access the modified feature, and (ii) a call to the first machine learning model to generate an outcome based on the modified feature.
9. A system comprising:
one or more computer processors; and
one or more computer-readable mediums storing instructions that, when executed by the one or more computer processors, cause the system to perform operations comprising:
accessing source data stored in a data store, the source data comprising raw data;
generating a feature based on the source data;
training a set of machine learning models based on the feature;
storing the feature in a feature store;
storing, in the feature store, metadata in association with the feature, the metadata including: (i) a mapping from the feature to upstream lineage data indicating the source data used to generate the feature, and (ii) a mapping from the feature to downstream lineage data indicating the set of machine learning models trained based on the feature;
determining, based on the metadata in the feature store including the mapping from the feature to the upstream lineage data, that a modification has been made to the feature;
in response to determining that the modification has been made to the feature, identifying, based on metadata in the feature store including the mapping from the feature to the downstream lineage data: (i) a first machine learning model from the set of machine learning models trained based on the feature and (ii) a model serving endpoint that deploys the first machine learning model; and
transmitting a notification to the model serving endpoint that deploys the first machine learning model, the notification indicating that the feature has been modified, wherein the model serving endpoint is a web service.
10. The system of claim 9 , the operations further comprising:
updating the mapping from the feature to the upstream lineage data based on the modification to the feature, wherein the mapping is updated to reflect modified upstream lineage data indicating revised source data.
11. The system of claim 9 , the operations further comprising:
generating the mapping from the feature to the upstream lineage data indicating the source data used to generate the feature.
12. The system of claim 9 , the operations further comprising:
determining that the feature is used to train a new machine learning model, the new machine learning model not in the set of machine learning models; and
updating the mapping from the feature to downstream lineage data indicating the new machine learning model.
13. The system of claim 9 , the operations further comprising:
storing the first machine learning model in a model store along with the mapping from the feature to the downstream lineage data indicating that the feature was used to train the first machine learning model, the model store storing at least a second machine learning model different than the first machine learning model.
14. The system of claim 9 , wherein the mapping from the feature to the upstream lineage data further comprises code used to determine the feature.
15. The system of claim 9 , the operations further comprising:
receiving data defining a second feature, the data defining the second feature identifying source data used to generate the second feature;
generating a mapping from the second feature to upstream lineage data indicating the source data used to generate the second feature; and
storing, in the feature store in association with the second feature, the mapping from the second feature to the upstream lineage data.
16. A non-transitory computer-readable medium storing instructions that, when executed by is one or more computer processors of one or more computing devices, cause the one or more computing devices to perform operations comprising:
accessing source data stored in a data store, the source data comprising raw data;
generating a feature based on the source data;
training a set of machine learning models based on the feature;
storing the feature in a feature store;
storing, in the feature store, metadata in association with the feature, the metadata including: (i) a mapping from the feature to upstream lineage data indicating the source data used to generate the feature, and (ii) a mapping from the feature to downstream lineage data indicating the set of machine learning models trained based on the feature;
determining, based on the metadata in the feature store including the mapping from the feature to the upstream lineage data, that a modification has been made to the feature;
in response to determining that the modification has been made to the feature, identifying, based on metadata in the feature store including the mapping from the feature to the downstream lineage data: (i) a first machine learning model from the set of machine learning models trained based on the feature and (ii) a model serving endpoint that deploys the first machine learning model; and
transmitting a notification to the model serving endpoint that deploys the first machine learning model, the notification indicating that the feature has been modified, wherein the model serving endpoint is a web service.
17. The non-transitory computer-readable medium of claim 16 , the operations further comprising:
updating the mapping from the feature to the upstream lineage data based on the modification to the feature, wherein the mapping is updated to reflect modified upstream lineage data indicating revised source data.
18. The non-transitory computer-readable medium of claim 16 , the operations further comprising:
generating the mapping from the feature to the upstream lineage data indicating the source data used to generate the feature.
19. The non-transitory computer-readable medium of claim 16 , the operations further comprising:
determining that the feature is used to train a new machine learning model, the new machine learning model not in the set of machine learning models; and
updating the mapping from the feature to downstream lineage data indicating the new machine learning model.
20. The non-transitory computer-readable medium of claim 16 , the operations further comprising:
storing the first machine learning model in a model store along with the mapping from the feature to the downstream lineage data indicating that the feature was used to train the first machine learning model, the model store storing at least a second machine learning model different than the first machine learning model.