Below is a selection of repositories related to my doctoral and master’s research work:
-
EHML: Extended Hybrid Machine Learning. Implementation of advanced extensions to hybrid machine learning frameworks, including physics-constrained data augmentation for multi-fidelity surrogate modeling. Developed using TensorFlow and integrated with Abaqus finite element simulations. Repository: https://github.com/shayansss/ehml
-
PSA: Pre-Stress Algorithm. A unified optimization framework for large-scale pre-stressing analysis in articular cartilage models. Implemented using Abaqus Fortran subroutines combined with Python-based automation scripts. Repository: https://github.com/shayansss/psa
-
HML: Hybrid Machine Learning. Implementation of a novel hybrid machine learning methodology for multi-fidelity surrogate modeling of finite element simulations, with applications in multi-physics modeling of soft biological tissues. Repository: https://github.com/shayansss/hml
-
PMSE: Pointwise Mean Squared Error. Development of a pointwise error metric tailored for machine-learning-based surrogate modeling. Implemented in Python using Keras and coupled with Abaqus simulations. Repository: https://github.com/shayansss/pmse
-
BioUMAT: Abaqus Fortran Subroutines for Cartilage Multiphase Modeling. Fortran 77 implementation of UMAT, FLOW, and SDVINI subroutines for a multiphasic cartilage model originally proposed in my Master’s thesis. With minor refinements, this model has been used in multiple peer-reviewed publications. Repository: https://github.com/shayansss/msc
-
PhD Dissertation. The full LaTeX source code of my doctoral dissertation is available here: Repository: https://github.com/shayansss/PhD
Due to confidentiality agreements related to my professional engagements, I am unable to share other datasets and proprietary code developed in the context of industry projects.