You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
hi lulu,
I have been working with the "hard_constraints_discontinuous_func.py" and "soft_constraints.py" files for the 2D linear elastic steady problem. However, but I am unable to obtain consistent results when running the code on my computer.
For instance, when running "soft_constraints.py" and inspecting the output file for the inverse problem, "2D_elastic_static_variables.dat," I discovered that the values of E and nu were [-8.04e-01, 1.69e-01], which do not match the expected/accurate values.
The text was updated successfully, but these errors were encountered:
Please see lines 159 and 160 of the code. The "2D_elastic_static_variables.dat" file does not directly output E and nu, instead it outputs a transformation of these predicted values. You can use the following equations to compute the true predicted values.
E = (tf.tanh(E_) + 1.0) * 3e5
nu = (tf.tanh(nu_) + 1.0) / 4
Please see section 2.4 of our paper if you have questions on why we use these transformations.
hi lulu,
I have been working with the "hard_constraints_discontinuous_func.py" and "soft_constraints.py" files for the 2D linear elastic steady problem. However, but I am unable to obtain consistent results when running the code on my computer.
For instance, when running "soft_constraints.py" and inspecting the output file for the inverse problem, "2D_elastic_static_variables.dat," I discovered that the values of E and nu were [-8.04e-01, 1.69e-01], which do not match the expected/accurate values.
The text was updated successfully, but these errors were encountered: