Transforming model data
Transforming model data, including: applying a transformation to data generated by a machine learning model; and providing the transformed data to a consumer of the transformed data, wherein the consumer is configured to: reverse the transformation applied to the transformed data; and perform one or more operations based on the data.
1 . A computer-implemented method, comprising:
applying, to an output of a machine learning model, a transformation, the transformation generating transformed data;
causing, by a consumer of the transformed data, a reversal of the transformation, the reversal of the transformation resulting in the output of the machine learning model; and
causing, by the consumer, performance of an operation on the output of the machine learning model.
2 . The computer-implemented method of claim 1 , wherein the transformation comprises an encryption operation.
3 . The computer-implemented method of claim 1 , wherein the transformation comprises an arithmetic transformation.
4 . The computer-implemented method of claim 1 , wherein the transformation comprises a Boolean operation.
5 . The computer-implemented method of claim 1 , wherein the machine learning model comprises a neural network and the transformation applied to the output of the machine learning model is performed by an output layer of the neural network.
6 . The computer-implemented method of claim 1 , further comprising:
submitting, to a relational data structure mapping an input key to a corresponding output value, a query based on location data;
receiving, in response to the query, input data for the machine learning model;
submitting the input data to the machine learning model; and
receiving, from the machine learning model, an output indicating one or more features associated with the location data.
7 . The computer-implemented method of claim 6 , wherein the relational data structure comprises a database mapping one or more locations to one or more values.
8 . The computer-implemented method of claim 7 , wherein the one or more values comprise one or more features corresponding to each of the one or more locations.
9 . The computer-implemented method of claim 6 , wherein the location data comprises location data associated with an autonomous vehicle.
10 . The computer-implemented method of claim 1 , wherein the machine learning model and the consumer are implemented in an autonomous vehicle.
11 . A computer-implemented method, comprising:
submitting, to a relational data structure mapping an input key to a corresponding output value, a query based on location data;
receiving, in response to the query, input data for a machine learning model;
submitting the input data to the machine learning model; and
receiving, from the machine learning model, an output indicating one or more features associated with the location data.
12 . The computer-implemented method of claim 11 , wherein the relational data structure comprises a database mapping one or more locations to one or more features corresponding to each of the one or more locations.
13 . The computer-implemented method of claim 12 , wherein the location data comprises location data associated with an autonomous vehicle, and wherein submitting the query and receiving the response to the query are performed by the autonomous vehicle.
14 . The computer-implemented method of claim 13 , wherein the database is implemented in the autonomous vehicle.
15 . The computer-implemented method of claim 13 , wherein the database is implemented remote to the autonomous vehicle.
16 . The computer-implemented method of claim 11 , further comprising:
applying, to an output of the machine learning model, a transformation, the transformation generating transformed data;
causing, by a consumer of the transformed data, a reversal of the transformation, the reversal of the transformation resulting in the output of the machine learning model; and
causing, by the consumer, performance of an operation on the output of the machine learning model.
17 . The computer-implemented method of claim 16 , wherein the machine learning model comprises a neural network and the transformation applied to the output of the machine learning model is performed by an output layer of the neural network.
18 . The computer-implemented method of claim 16 , wherein the machine learning model and the consumer are implemented in an autonomous vehicle.
19 . An apparatus, the apparatus configured to perform steps comprising:
applying, to an output of a machine learning model, a transformation, the transformation generating transformed data;
causing, by a consumer of the transformed data, a reversal of the transformation, the reversal of the transformation resulting in the output of the machine learning model; and
causing, by the consumer, performance of an operation on the output of the machine learning model.
20 . The apparatus of claim 19 , wherein the steps further comprise:
submitting, to a relational data structure mapping an input key to a corresponding output value, a query based on location data;
receiving, in response to the query, input data for the machine learning model;
submitting the input data to the machine learning model; and
receiving, from the machine learning model, an output indicating one or more features associated with the location data.