David Donahue

Thank you for the code! By default, Pytorch linear layers use uniform Kaiming initialization, which is meant to be used before a ReLU activation with uniform [+1, -1] inputs. However,...

ActNorm reinitializes on model load

3

Thank you for the repo! Upon reloading the Glow model from save after training, cross entropy performance diminished. This was likely because the ActNorm module was being reinitialized on the...

# zero-initialize the last layer, as in the paper if n_hidden > 0: init.constant(self.layers[-1].weight, 0) else: init.constant(self.input_trigger.weight, 0) if context_dim is not None: init.constant(self.input_context.weight, 0) The BasicSynthesizer class sets all...

Debug panel fields not displaying correctly

7

The "Variables" and "Call stack" fields in the Breakpoint panel consume everything below them when clicked. Steps to reproduce: - Start cursor - Go to debug panel - Open "Variables"...