photontorch
photontorch copied to clipboard
flaport
Highly parallel simulation and optimization of photonic circuits in time and frequency domain based on the deep-learning framework PyTorch
A lot of memory usage can probably be avoided when transitioning to sparse tensor representations under the hood.
As long as PyTorch does not sufficiently support complex numbers out of the box, complex number support won't be added to Photontorch. This issue is here to track the progress...
When I tried to run # create simulation environment freq_env = pt.Environment( wl=1e-6*np.linspace(1.45, 1.65, 1000), freqdomain=True ) with freq_env: # perform simulation detected = circuit(source=1) # constant source with amplitude...
