ooDACE Toolbox: A Flexible Object-Oriented Kriging Implementation
Authors: Ivo Couckuyt, Tom Dhaene, Piet Demeester
JMLR 2014 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | To assess the performance of the oo DACE toolbox a comparison between the oo DACE toolbox and the DACE toolbox1 is performed using the 2D Branin function. To that end, 20 data sets of increasing size are constructed, each drawn from an uniform random distribution. The number of observations ranges from 10 to 200 samples with steps of 10 samples. For each data set, a DACE toolbox1 model, a oo DACE ordinary Kriging and a oo DACE blind Kriging model have been constructed and the accuracy is measured on a dense test set using the Average Euclidean Error (AEE). Moreover, each test is repeated 1000 times to remove any random factor, hence the average accuracy of all repetitions is used. Results are shown in Figure 2a. |
| Researcher Affiliation | Academia | Ivo Couckuyt EMAIL Tom Dhaene EMAIL Piet Demeester EMAIL Ghent University i Minds Department of Information Technology (INTEC) Gaston Crommenlaan 8 9050 Gent, Belgium |
| Pseudocode | No | The paper describes the implementation details and class structure but does not present any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | The oo DACE toolbox is an object-oriented Matlab toolbox implementing a variety of Kriging flavors and extensions. [...] Usage instructions, design documentation, and stable releases can be found at http://sumo.intec.ugent.be/?q=oo DACE. |
| Open Datasets | No | To that end, 20 data sets of increasing size are constructed, each drawn from an uniform random distribution. The paper describes generating data from the 2D Branin function and a uniform random distribution, rather than providing access information for a pre-existing public dataset. |
| Dataset Splits | No | For each data set, a DACE toolbox1 model, a oo DACE ordinary Kriging and a oo DACE blind Kriging model have been constructed and the accuracy is measured on a dense test set using the Average Euclidean Error (AEE). While a 'dense test set' is mentioned, specific details about training/validation/test splits, such as percentages or sample counts, are not provided. |
| Hardware Specification | No | The paper does not provide specific details regarding the hardware used for running experiments. |
| Software Dependencies | No | The paper states that the 'oo DACE toolbox is an object-oriented Matlab toolbox' but does not specify the version of Matlab or any other software dependencies with version numbers. |
| Experiment Setup | No | The paper describes the setup for comparing the toolboxes, including data generation, sample sizes, and evaluation metric. However, it does not specify concrete hyperparameter values or detailed training configurations for the Kriging models themselves. |