exporting generator and discriminator models into a .pth file

[arnav-newzera]

i trained my dataset on the recipe given in https://github.com/jik876/hifi-gan . In the training checkpoints folder i see do_XXXX and g_XXXX (where XXXX are the step numbers ) files which i assume are discriminator and generator models. Now the question is: how do i export these into a model.pth file also generate a config.json file for the same . so that i can use it with other accoustic models?