Running notebook 03a_generate_synthetic_data.ipynb: What parameters to use for RWW PyTorch Model

[dChakr]

Hi there!

My name is Dyuti Chakraborty, I'm a final year undergraduate student at Imperial College London (UK) and I'm undertaking a master's project in whole-brain modelling.

I came across the paper "Deep Learning-Based Parameter Estimation for Neurophysiological Models of Neuroimaging Data" (doi: https://doi.org/10.1101/2022.05.19.492664) and wanted to try out the code mentioned. I tried to run one of the examples (notebooks/03a_generate_synthetic_data.ipynb), however came across some issues which I hoping could be cleared up!

I was having trouble running some functions, such as initiating the RNNWWD model in block 6, since some of the required arguments were missing. These arguments include values for some required parameters, such as gEE, gIE, gEI, etc. I have tried to use other values for these parameters, based on the paper or other examples, but am having difficulty reproducing your results. Other cases include missing the cost value and learning rate when calling the Model_fitting() function and the values of x0 and he0 when testing the model in F.test(). I was wondering whether you could provide the values of these parameters, or point to where they are mentioned in the literature?

Thanks!
Dyuti