z.q.mao
z.q.mao
@Kyubyong !Can you share the format of training corpus for training CRF models? thank you!
@fengxqinx thanks!
Hi @bfs18 ! the training process is so slowly ! Can you upload one copy?
@bfs18 ! thanks!
hi @bfs18 ! Nice. How many steps have been trained to get that results?
@bfs18 how about the teacher network's performance? the good teacher network is very very important!
Thank you for your - new PQMF was not resolved this problem!, - Breakpoints appear randomly - increasing stacks i will try!
hi @kan-bayashi ! increasing stacks not resolved this problem! I suspect that the deconvolution kernel is responsible for this problem!
@kan-bayashi Thanks!I will try this!
@zpcoftts! I try to change the size of convolution kernel, deepen the number of stack layers, modify the discriminant function, remove the weight normalize, etc., which can not solve this...