Method and apparatus for determining device information, and system
The present disclosure relates to methods and apparatus for determining device information. In one example method, an inference device sends a first request to a service discovery entity. The first request is used to request information about one or more training devices, and the first request includes an algorithm type or an algorithm identifier of a first model requested by the inference device. The inference device receives the information about the one or more training devices from the service discovery entity. The information about the one or more training devices includes capability information. The inference device determines a first training device from the information about the one or more training devices based on a preset condition.
1. A method for determining device information, comprising:
sending, by an inference device, a first request to a service discovery entity, wherein the first request requests information about one or more training devices included in the service discovery entity, and the first request comprises an algorithm type or an algorithm identifier of a first model requested by the inference device;
receiving, by the inference device, the information about the one or more training devices from the service discovery entity, wherein the information about the one or more training devices comprises capability information, and the information about the one or more training devices is determined by the service discovery entity based on the algorithm type or the algorithm identifier of the first model requested by the inference device;
determining, by the inference device, a first training device from the information about the one or more training devices based on a preset condition; and
sending, by the inference device, a third request to the service discovery entity, wherein the third request requests to register or update the information about the first model, and the third request comprises any one or more of the following information:
an analytics ID corresponding to the first model, an effective area, a service area, or a coverage area of an analytics corresponding to the first model.
2. The method according to claim 1 , wherein the capability information comprises any one or more of an algorithm type, an algorithm identifier, an algorithm performance evaluation indicator, algorithm convergence time, an algorithm convergence speed, or algorithm confidence.
3. The method according to claim 1 , wherein the preset condition comprises any one or more of the following:
algorithm convergence time corresponding to an algorithm of the first training device is the fastest in algorithm convergence time corresponding to an algorithm of any one of the one or more training devices; or
an algorithm performance evaluation indicator corresponding to an algorithm of the first training device is the highest in an algorithm performance evaluation indicator corresponding to an algorithm of any one of the training devices.
4. The method according to claim 3 , wherein the information about the one or more training devices further comprises any one of location information or load information, and the preset condition further comprises:
load of the first training device is the lowest in load of the one or more training devices.
5. The method according to claim 1 , wherein the method further comprises:
sending, by the inference device, a fourth request to the first training device, wherein the fourth request requests information about the first model, and the fourth request comprises any one or more of the following information corresponding to the first model: the algorithm type, the algorithm identifier, an algorithm performance requirement, a data type, or a data address; and
receiving, by the inference device, the information about the first model from the first training device, wherein the information about the first model comprises any one or more of a model identifier, a model input, or a model parameter.
6. The method according to claim 5 , wherein the information about the first model further comprises any one or more of a model output, additional model information, or a result of a model evaluation indicator.
7. A communication apparatus, comprising at least one processor and a memory, wherein the memory stores computer instructions for execution by the at least one processor to cause the communication apparatus to:
send a first request to a service discovery entity, wherein the first request requests information about one or more training devices included in the service discovery entity, and the first request comprises an algorithm type or an algorithm identifier of a first model requested by the communication apparatus;
receive the information about the one or more training devices from the service discovery entity, wherein the information about the one or more training devices comprises capability information, and the information about the one or more training devices is determined by the service discovery entity based on the algorithm type or the algorithm identifier of the first model requested by the communication apparatus;
determine a first training device from the information about the one or more training devices based on a preset condition; and
sending a third request to the service discovery entity, wherein the third request requests to register or update the information about the first model, and the third request comprises any one or more of the following information:
an analytics ID corresponding to the first model, an effective area, a service area, or a coverage area of an analytics corresponding to the first model.
8. The communication apparatus according to claim 7 , wherein the capability information comprises any one or more of an algorithm type, an algorithm identifier, an algorithm performance evaluation indicator, algorithm convergence time, an algorithm convergence speed, or algorithm confidence.
9. The communication apparatus according to claim 7 , wherein the preset condition comprises any one or more of the following:
algorithm convergence time corresponding to an algorithm of the first training device is the fastest in algorithm convergence time corresponding to an algorithm of any one of the one or more training devices; or
an algorithm performance evaluation indicator corresponding to an algorithm of the first training device is the highest in an algorithm performance evaluation indicator corresponding to an algorithm of any one of the training devices.
10. The communication apparatus according to claim 9 , wherein the information about the one or more training devices further comprises any one of location information or load information, and the preset condition further comprise:
load of the first training device is the lowest in load of the one or more training devices.
11. The communication apparatus according to claim 7 , wherein the computer instructions are for execution by the at least one processor to cause the communication apparatus to:
send a fourth request to the first training device, wherein the fourth request requests information about the first model, and the fourth request comprises any one or more of the following information corresponding to the first model: the algorithm type, the algorithm identifier, an algorithm performance requirement, a data type, or a data address; and
receive the information about the first model from the first training device, wherein the information about the first model comprises any one or more of a model identifier, a model input, or a model parameter.
12. The communication apparatus according to claim 11 , wherein the information about the first model further comprises any one or more of a model output, additional model information, or a result of a model evaluation indicator.
13. A communication system, comprising:
an inference device, comprising: at least one processor and a memory, wherein the memory stores computer instructions for execution by the at least one processor to cause the inference device to:
send a first request to a service discovery entity, wherein the first request requests information about one or more training devices included in the service discovery entity, and the first request comprises an algorithm type or an algorithm identifier of a first model requested by the inference device;
receive the information about the one or more training devices from the service discovery entity, wherein the information about the one or more training devices comprises capability information, and the information about the one or more training devices is determined by the service discovery entity based on the algorithm type or the algorithm identifier of the first model requested by the inference device;
determine a first training device from the information about the one or more training devices based on a preset condition; and
send a third request to the service discovery entity, wherein the third request requests to register or update the information about the first model, and the third request comprises any one or more of the following information:
an analytics ID corresponding to the first model, an effective area, a service area, or a coverage area of an analytics corresponding to the first model; and
the service discovery entity, configured to receive the first request from the inference device.
14. The communication system according to claim 13 , wherein the capability information comprises any one or more of an algorithm type, an algorithm identifier, an algorithm performance evaluation indicator, algorithm convergence time, an algorithm convergence speed, or algorithm confidence.
15. The communication system according to claim 13 , wherein the preset condition comprises any one or more of the following:
algorithm convergence time corresponding to an algorithm of the first training device is the fastest in algorithm convergence time corresponding to an algorithm of any one of the one or more training devices; or
an algorithm performance evaluation indicator corresponding to an algorithm of the first training device is the highest in an algorithm performance evaluation indicator corresponding to an algorithm of any one of the training devices.
16. The communication system according to claim 15 , wherein the information about the one or more training devices further comprises any one of location information or load information, and the preset condition further comprise:
load of the first training device is the lowest in load of the one or more training devices.
17. The communication system according to claim 13 , wherein the computer instructions are for execution by the at least one processor to cause the inference device to:
send a fourth request to the first training device, wherein the fourth request requests information about the first model, and the fourth request comprises any one or more of the following information corresponding to the first model: the algorithm type, the algorithm identifier, an algorithm performance requirement, a data type, or a data address; and
receive the information about the first model from the first training device, wherein the information about the first model comprises any one or more of a model identifier, a model input, or a model parameter.