1 to 10 of 152,602 Results
May 26, 2026 - Institute for Modelling and Simulation of Biomechanical Systems
Rosin, David, 2026, "Data for: PerSiVal: deep neural networks for pervasive simulation of an activation-driven continuum-mechanical upper limb model", https://doi.org/10.18419/DARUS-3795, DaRUS, V1
This dataset contains data needed to reproduce the results of the related publication on neural-network-based real-time prediction of surface deformation of the biceps under active contraction. This includes deep learning training data, sparse grid support data and the exact neural network models used in the publication. The architecture of said ne... |
May 26, 2026 -
Data for: PerSiVal: deep neural networks for pervasive simulation of an activation-driven continuum-mechanical upper limb model
Python Source Code - 2.9 KB -
MD5: 33b3ad607faa81f028ea2dc7ce76e0be
Script to combine the three separately trained networks for the coordinate components into one, adding the reference coordinates to the bias of the output layer in the process, to convert from deformation to Cartesian coordinates. |
May 26, 2026 -
Data for: PerSiVal: deep neural networks for pervasive simulation of an activation-driven continuum-mechanical upper limb model
Python Source Code - 1.3 KB -
MD5: 20d942d0689d903eb57295120b8912a5
Utility script containing all the code needed for some basic cross validation training. |
May 26, 2026 -
Data for: PerSiVal: deep neural networks for pervasive simulation of an activation-driven continuum-mechanical upper limb model
Python Source Code - 3.5 KB -
MD5: 0a47f1396b6027c593a257941b3b6439
Script file containing all the code needed to evaluate the sparse grid surrogate based on the data provided in this dataset. |
May 26, 2026 -
Data for: PerSiVal: deep neural networks for pervasive simulation of an activation-driven continuum-mechanical upper limb model
Unknown - 337.3 KB -
MD5: de2ed2b6631bd880c8e473158e2f8060
Content of the neural network model predicting muscle surface coordinates based on muscle activation. Needs to be inside a folder representing the entire TensorFlow model. |
May 26, 2026 -
Data for: PerSiVal: deep neural networks for pervasive simulation of an activation-driven continuum-mechanical upper limb model
Unknown - 140.4 KB -
MD5: 2c6ff93e346677b9ea1955e8f3ef2aa6
Data representing a sparse grid surrogate, taking activation levels as input and outputting the corresponding elbow joint angle from the FE model. |
May 26, 2026 -
Data for: PerSiVal: deep neural networks for pervasive simulation of an activation-driven continuum-mechanical upper limb model
Python Source Code - 1.4 KB -
MD5: 9147e8480bf4b3163bee6f6bdc4442dd
Script explaining how to evaluate the sparse grid surrogate based on the data provided in this dataset. |
May 26, 2026 -
Data for: PerSiVal: deep neural networks for pervasive simulation of an activation-driven continuum-mechanical upper limb model
Unknown - 135.5 MB -
MD5: 726bfd368a99ac024441e4a3cb27720b
Pickled Numpy-file containing the support for the sparse grid interpolation of the Cartesian coordinates of each finite element mesh-node on the surface of the biceps'. |
May 26, 2026 -
Data for: PerSiVal: deep neural networks for pervasive simulation of an activation-driven continuum-mechanical upper limb model
Unknown - 112.1 MB -
MD5: 22dd49433006aa98f56eae5cc78d79b4
Numpy array containing the training data for the neural network model (net_acts2coords_biceps) as input-output sample pairs. Each line contains muscle activation input in the first five columns, followed by 2809 x coordinates, then just as many y coordinates, then z coordinates. |
May 26, 2026 -
Data for: PerSiVal: deep neural networks for pervasive simulation of an activation-driven continuum-mechanical upper limb model
Python Source Code - 2.1 KB -
MD5: 880e0051ff4bc3d8f2d017446c394ad0
Script to test the combined neural network model for the worst case error on randomly picked activation vectors. |
