This is a program intended to consolidate quantitative simulation and experimental data from MFH and MPS. A machine learning algorithm is used to extrapolate predictions for numerical values, such as scaled SLP values (for MFH), and harmonic amplitudes (for MPS).
Download the relevant pkl files (or zip files containing pkl files). The relevant models are:
slpmodel: predicts SLP values from simulation data
slpmodel-exp: predicts SLP values from experimental data
mpsmodel-ratio: predicts ratio of 5th:3rd MPS harmonic amplitudes from simulation data
mpsmodel-ratio-frozen: predicts ratio of 5th:3rd MPS harmonic amplitudes from simulation data with Brownian relaxation off
mpsmodel-tau: predicts relaxation time from MPS simulation data
mpsmodel-tau-frozen: predicts relaxation time from MPS simulation data with Brownian relaxation off
mpsmodel-tan: predicts phase lag (omega x tau) from MPS simulation data
mpsmodel-tan-frozen: predicts phase lag (omega x tau) from MPS simulation data with Brownian relaxation off
Run the relevant prediction notebook (slp_predict_from_sim, slp_predict_from_exp, mps_predict_from_sim).