Identifiability of the stress exponent in weighted multidimensional scaling: results archive and code snapshot
收藏资源简介:
Current snapshot of the actual source code and data results v1.1.2 Measured results and the exact code snapshot behind the manuscript"Identifiability of the stress exponent in weighted multidimensional scaling". The archive contains two files: 1. Sammon_data_<date>.zip - the measured experimental results as CSV files (results/data/). Every number, table and figure in the manuscript and its supplement is derived from these files by the published scripts. 2. stress-exponent-identifiability-<tag>.zip - a snapshot of the source code at the version submitted to the journal, identical to the tag of the same name in the GitHub repository. Input datasets are not included: they are either downloaded from publicrepositories (scikit-learn, OpenML, SNAP, SocioPatterns) or generateddeterministically from the seeds in the configuration files, in both cases bythe scripts in the code archive. Derived artefacts (figures, tables, numbermacros, embeddings) are not included either, because they are regenerated fromthe CSV files by a single command; see README.md in the code archive. Code is released under the MIT license.



