Intelligent execution of compute intensive numerical simulation models
In a method for intelligently executing predictive simulator, a processor may input a previous input vector of conditions for a predictive simulator collected at a first time into a machine-learning (ML) model. A processor may input a current input vector of conditions for the predictive simulator collected at a second time into the ML model. A processor may determine using the ML model, a binary similarity index. The binary similarity index represents a prediction of similarity between a first output from the predictive simulator based on the previous input and a second output from the predictive simulator based on the current input.
1. A computer-implemented method for intelligently executing predictive simulator, comprising:
inputting, by one or more processors, a previous input vector into a machine-learning (ML) model, wherein the previous input vector comprises first conditions for producing a first output from a predictive simulator for a first time;
inputting a current input vector into the ML model, wherein the current input vector comprises second conditions for producing a second output from the predictive simulator for a second time;
determining, using the ML model, a binary similarity index, wherein the binary similarity index represents a prediction of similarity between the first output and the second output.
2. The computer-implemented method of claim 1 , further comprising inputting exogeneous inputs, wherein the exogenous inputs comprise a selection from the group consisting of: location-based metrics, outputs of the predictive simulator, and use metrics of the predictive simulator.
3. The computer-implemented method of claim 1 , wherein the current input vector comprises a selection from the group consisting of: initial conditions, boundary conditions, dynamic fields, fixed fields, and observed environmental conditions such as precipitation, wind, and temperature.
4. The computer-implemented method of claim 1 , wherein the previous input vector comprises a selection from the group consisting of: initial conditions, boundary conditions, dynamic fields, fixed fields, and observed environmental conditions such as precipitation, wind, and temperature.
5. The computer-implemented method of claim 1 , wherein the binary similarity index comprises:
a user similarity component comprising a forecasting requirement of a user; and
a model similarity component comprising a clustering of the previous input vector and the current input vector.
6. The computer-implemented method of claim 1 , further comprising running the current input vector through the predictive simulator only if the binary similarity index indicates a TRUE similarity.
7. The computer-implemented method of claim 1 , further comprising training the ML model, wherein training the ML model comprises inputting pairs comprising: (i) inputs to the predictive simulator and (ii) resulting outputs from the predictive simulator.
8. The computer-implemented method of claim 7 , wherein the pairs of inputs are input based on a selection from the group consisting of: seasonality, topography, and other hyper-parameter-specific partition.
9. A computer program product for training temporal precipitation interpolation models, comprising:
one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions comprising:
program instructions to input a previous input vector into a machine-learning (ML) model, wherein the previous input vector comprises first conditions for producing a first result from a predictive simulator for a first time;
program instructions to input a current input vector into the ML model, wherein the current input vector comprises second conditions for producing a second result from the predictive simulator for a second time;
program instructions to determine, using the ML model, a binary similarity index, wherein the binary similarity index represents a prediction of similarity between the first output and the second output.
10. The computer program product of claim 9 , further comprising program instructions to input exogeneous inputs, wherein the exogenous inputs comprise a selection from the group consisting of: location-based metrics, outputs of the predictive simulator, and use metrics of the predictive simulator.
11. The computer program product of claim 9 , wherein the current input vector comprises a selection from the group consisting of: initial conditions, boundary conditions, dynamic fields, fixed fields, and observed environmental conditions such as precipitation, wind, and temperature.
12. The computer program product of claim 9 , wherein the previous input vector comprises a selection from the group consisting of: initial conditions, boundary conditions, dynamic fields, fixed fields, and observed environmental conditions such as precipitation, wind, and temperature.
13. The computer program product of claim 9 , wherein the binary similarity index comprises:
a user similarity component comprising a forecasting requirement of a user; and
a model similarity component comprising a clustering of the previous input vector and the current input vector.
14. The computer program product of claim 9 , further comprising program instructions to run the current input vector through the predictive simulator only if the binary similarity index indicates a TRUE similarity.
15. The computer program product of claim 9 , further comprising program instructions to train the ML model, wherein training the ML model comprises inputting pairs comprising: (i) inputs to the predictive simulator and (ii) resulting outputs from the predictive simulator.
16. The computer program product of claim 15 , wherein the pairs of inputs are input based on a selection from the group consisting of: seasonality, topography, and other hyper-parameter-specific partition.
17. A computer system for training temporal precipitation interpolation models, comprising:
one or more computer processors, one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media for execution by at least one of the one or more computer processors, the program instructions comprising:
program instructions to input a previous input vector into a machine-learning (ML) model, wherein the previous input vector comprises first conditions for producing a first result from a predictive simulator for a first time;
program instructions to input a current input vector into the ML model, wherein the current input vector comprises second conditions for producing a second result from the predictive simulator for a second time;
program instructions to determine, using the ML model, a binary similarity index, wherein the binary similarity index represents a prediction of similarity between the first output and the second output.
18. The computer system of claim 17 , further comprising program instructions to input exogeneous inputs, wherein the exogenous inputs comprise a selection from the group consisting of: location-based metrics, outputs of the predictive simulator, and use metrics of the predictive simulator.
19. The computer system of claim 17 , further comprising program instructions to run the current input vector through the predictive simulator only if the binary similarity index indicates a TRUE similarity.
20. The computer system of claim 17 , wherein the binary similarity index comprises:
a user similarity component comprising a forecasting requirement of a user; and
a model similarity component comprising a clustering of the previous input vector and the current input vector.