Katsuhito Sudoh
Katsuhito Sudoh
When I implemented stacked LSTMs, there was a need to maintain layer-wise parameter matrices in one Parameter, such as - Wx[0]: Parameter Wx for layer 1, - Wx[1]: Parameter Wx...
Hi, I've just started to use SyMGIZA++, and found a problem in mergeAlignment.cpp. It sometimes fails to retrieve a few sentences in the end of dataset but correctly generated in...
I faced a problem on very short inputs. If the training data contain a sentence with just one element like: XXX. tree2seq trainer has failed by segmentation fault.