Spatially Explicit Hyperparameter Optimization for Neural Networks by Minrui Zheng

Author:Minrui Zheng
Language: eng
Format: epub, pdf
ISBN: 9789811653995
Publisher: Springer Singapore

Specifically, we chose one of the groups as our experimental dataset. In this case study, group 4 with 9254 records (parcels here) was selected: 6478 records (70%) were assigned as the training set, and the remaining 2776 records (30%) for testing.

We used the same training dataset to fit the linear regression model and the same testing dataset was used to predict values based on the fitted linear model. The ordinary least squares model was applied in this study. Table 4.3 summarizes the results of the linear regression model. The R2 of the linear model is 0.7183, and the predicted MSE based on the testing dataset is 0.0068. In the ANN model, the MSE of the optimal hyperparameter set is around 0.004 for both hyperparameter methods (reported in detail next). These results suggest that the ANN model has a better generalization performance than linear regression modeling.Table 4.3Results of linear regression modeling (see Table 4.2 for definitions of these variables)

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.