Execution time prediction on FastDeepIoT Profiling Dataset Galaxy Nexus LSTM layer (test)
2.9MAPEFastDeepIoT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| FastDeepIoTRegression Model=Tree-structured linear regression, Input Feature Vector=x2018.09 | 2.9 | 1.4 | 0.997 | |
| DNNRegression Model=multilayer perceptron, Input Feature Vector=x2018.09 | 2.9 | 1.3 | 0.997 | |
| GBRTRegression Model=gradient boosted regression trees, Input Feature Vector=x2018.09 | 6 | 2.7 | 0.987 | |
| RFRegression Model=random forest regression, Input Feature Vector=x2018.09 | 7.8 | 3.3 | 0.985 | |
| DTRegression Model=classification and regression trees, Input Feature Vector=x2018.09 | 8.4 | 3 | 0.983 | |
| SVRRegression Model=support vector regression with RBF kernel, Input Feature Vector=x2018.09 | 66.8 | 26.2 | -0.196 |