General purpose input/output data capture and neural cache system for autonomous machines

1. An apparatus comprising:

one or more processors to:

capture, via an image capturing device, one or more images of one or more objects, wherein the one or more images represent input data associated with a neural network;

determine accuracy of first output results generated by a default neural caching system by comparing the first output results with second output results predicted by a custom neural caching system; and

output, based on the accuracy, a final output results including at least one of the first output results or the second output results, wherein one or more components of at least one of the default neural caching system and the custom neural caching system are turned on or off based on the accuracy,

wherein the default neural caching system is turned off when the predicted second output results exceed an accuracy threshold, and wherein the custom neural caching system is turned off when the predicted second output results fail to exceed the accuracy threshold, wherein the accuracy threshold equals the generated first output results.

2. The apparatus of claim 1 , wherein the one or more processors to facilitate:

provide the input data to the custom neural caching system for processing, wherein the input data is separated by the input data forker for caching purposes and training purposes; and

join the first and second output results to be compared.

3. The apparatus of claim 1 , wherein the one or more processors to:

associate at least one of a tag-pair and a dependency tag to the input data and the first output results; and

capture the input data from the input data forker, and receive one or more of the tag-pair, the dependency tag, and the first output results.

4. The apparatus of claim 3 , wherein the one or more processors to receive one or more of the input data, the tag-pair, the dependency tag, and the first output results, and to store one or more of the input data, the tag-pair, the dependency tag, and the first output results.

5. The apparatus of claim 1 , wherein the one or more processors to train, based on the accuracy, the default neural caching system to facilitate a trained neural model to predict the second output results.

6. The apparatus of claim 1 , wherein the input capturing device comprises at least one of one or more cameras, one or more robot eyes, one or more microphones, and one or more sensors, wherein the apparatus comprises an autonomous machine or an artificially intelligent agent, wherein the autonomous machine includes at least one of one or more robots, one or more self-driving vehicles, and one or more self-operating equipment.

7. A method comprising:

capturing, via an image capturing device, one or more images of one or more objects, wherein the one or more images represent input data associated with a neural network;

determining accuracy of first output results generated by a default neural caching system by comparing the first output results with second output results predicted by a custom neural caching system; and

outputting, based on the accuracy, a final output results including at least one of the first output results or the second output results, wherein one or more components of at least one of the default neural caching system and the custom neural caching system are turned on or off based on the accuracy,

wherein the default neural caching system is turned off when the predicted second output results exceed an accuracy threshold, and wherein the custom neural caching system is turned off when the predicted second output results fail to exceed the accuracy threshold, wherein the accuracy threshold equals the generated first output results.

8. The method of claim 7 , further comprising:

providing the input data to the custom neural caching system for processing, wherein the input data is separated by the input data forker for caching purposes and training purposes; and joining the first and second output results to be compared.

9. The method of claim 7 , further comprising:

associating at least one of a tag-pair and a dependency tag to the input data and the first output results;

capturing the input data from the input data forker; and

receiving one or more of the tag-pair, the dependency tag, and the first output results.

10. The method of claim 9 , further comprising receiving one or more of the input data, the tag-pair, the dependency tag, and the first output results, and storing one or more of the input data, the tag-pair, the dependency tag, and the first output results.

11. The method of claim 7 , further comprising training, based on the accuracy, the default neural caching system to facilitate a trained neural model to predict the second output results.

12. The method of claim 7 , wherein the input capturing device comprises at least one of one or more cameras, one or more robot eyes, one or more microphones, and one or more sensors, wherein the apparatus comprises an autonomous machine or an artificially intelligent agent, wherein the autonomous machine includes at least one of one or more robots, one or more self-driving vehicles, and one or more self-operating equipment.

13. At least one non-transitory machine-readable medium comprising a plurality of instructions, when executed on a computing device, to facilitate operations comprising:

capturing, via an image capturing device, one or more images of one or more objects, wherein the one or more images represent input data associated with a neural network;

determining accuracy of first output results generated by a default neural caching system by comparing the first output results with second output results predicted by a custom neural caching system; and

outputting, based on the accuracy, a final output results including at least one of the first output results or the second output results, wherein one or more components of at least one of the default neural caching system and the custom neural caching system are turned on or off based on the accuracy,

wherein the default neural caching system is turned off when the predicted second output results exceed an accuracy threshold, and wherein the custom neural caching system is turned off when the predicted second output results fail to exceed the accuracy threshold, wherein the accuracy threshold equals the generated first output results.

14. The machine-readable medium of claim 13 , wherein the method further comprises:

providing the input data to the custom neural caching system for processing, wherein the input data is separated by the input data forker for caching purposes and training purposes; and joining the first and second output results to be compared.