This record contains the simulated Zernike-coefficient datasets, source code, model configurations, pretrained weights, and machine-readable numerical results supporting the SI-Mamba study. The archived learning models are LSTM, Mamba, and SI-Mamba. It includes the training set, five held-out test sets, 25 boundary-condition datasets, evaluation and boundary-analysis scripts, and the exact software environment. Figure-generation code and figures are intentionally excluded. Source code is licensed under Apache-2.0; data, pretrained weights, and numerical results are licensed under CC BY 4.0. See the included README and SHA256SUMS for reproduction instructions and integrity verification.
qin xiao· Zenodo (CERN European Organi...· 0 citations
This record contains the simulated Zernike-coefficient datasets, source code, model configurations, pretrained weights, and machine-readable numerical results supporting the SI-Mamba study. The archived learning models are LSTM, Mamba, and SI-Mamba. It includes the training set, five held-out test sets, 25 boundary-condition datasets, evaluation and boundary-analysis scripts, and the exact software environment. Figure-generation code and figures are intentionally excluded. Source code is licensed under Apache-2.0; data, pretrained weights, and numerical results are licensed under CC BY 4.0. See the included README and SHA256SUMS for reproduction instructions and integrity verification.
qin xiao· Zenodo (CERN European Organi...· 0 citations