RNNSharp is a toolkit of deep recurrent neural network which is widely used for many different kinds of tasks, such as sequence labeling, sequence-to-sequence and so on. It's written by C# language and based on .NET framework 4.6 or above versions. RNNSharp supports many different types of networks, such as forward and bi-directional network, sequence-to-sequence network, and different types of layers, such as LSTM, Softmax, sampled Softmax and others.
when i training a pos tagger using RNN,CRF,RNN+CRF,LSTM+CRF,the result is very good, but LSTM is not better than others,I have same training, development and test corpus, what is my problem?
Thanks @Bilgeyar for pointing it out. It's a bug when computing output layer in forward pass. I will fix it today. Once's the fixing is checked-in, I will update this thread.
zhongkaifu
commented
7 years ago
when i training a pos tagger using RNN,CRF,RNN+CRF,LSTM+CRF,the result is very good, but LSTM is not better than others,I have same training, development and test corpus, what is my problem?