whre is the sep_module (calss or funtion）in sourcesep.py

[xiaoyiming]

Really nice job!!!， I found that “sep_module”（class or function ) was used in sourcesep.py file.
However, I could not find the definition. the “sep_module” was used as followed:

"
spec_mix, phase_mix = sep_module(pr).stft(samples_trunc[:, :, 0], pr)
spec_mix = crop_spec(spec_mix)
phase_mix = crop_spec(phase_mix)

    self.specgram_op, phase = map(crop_spec, sep_module(pr).stft(samples_trunc[:, :, 0], pr))
    self.auto_op = sep_module(pr).istft(self.specgram_op, phase, pr)

    self.net = sep_module(pr).make_net(
      self.ims_ph, samples_trunc, spec_mix, phase_mix, 
      pr, reuse = False, train = False) "

[Andrew Owens]

Oops, sep_module() should simply return the "sourcesep" module (the code is written this way to support baselines, but we haven't released those). Thanks for catching that!

[xiaoyiming]

Thanks for reply. However, I found another problem as follows:
def slim_losses_with_prefix(prefix, show = True):
losses = tf.losses.get_regularization_losses()
losses = [x for x in losses if prefix is None or x.name.startswith(prefix)]
if show:
print 'Collecting losses for prefix %s:' % prefix
for x in losses:
print x.name
print
return mu.maybe_add_n(losses)
where the losses is always empty?

[Bin Duan]

@xiaoyiming Hello, I notice the miss definition too. Could you please share the way you deal with the undefined "sourcesep" module? Thank you.