-
For OCR postcorrection, `TextLine.Word.Glyph.TextEquiv` can be more valueable than just `TextLine.TextEquiv`. It allows to build up a lattice (or rather, confusion network) of alternative character hy…
-
您好!请问在Lattice LSTM中用的是单向LSTM吗?或者说是通过改进LSTM-CRF模型得到的Lattice LSTM模型的吗?虚心请教。
-
请问acl 2018 论文Chinese NER Using Lattice LSTM里正文的中文是怎么在不影响acl模板格式的情况下打出来的呢?网上查到的方法都会改变原有的模板格式。
谢谢。
-
您好!
想请教一下论文里面的Lattice LSTM模型有提到是用的双向还是单向LSTM吗?希望您解答,谢谢。
-
As the title shows, if you want to train LatticeLSTM by mini-batch, please refer to [Batch-Parallel-LatticeLSTM](https://github.com/LeeSureman/Batch_Parallel_LatticeLSTM)
-
Hi,
Thanks for your nice work.
By the way, I want to know if this implementation of 3D version can be directly used for irregular discrete points?
-
I would like to replace the `layer_lstm()` with a multiplicative LSTM implementation (e.g. from [here](https://github.com/titu1994/Keras-Multiplicative-LSTM)). I'm not sure which files I have to chang…
-
作者您好:
我用weibo数据集(BMEOS模式)在您的原始代码(参数等均未改动)上跑了lattice model的实验,在NE,NM,ALL测试集上的F值都没有达到您论文汇报的结果(分别为:51.77vs53.04,60.00vs62.25,56.00vs58.79)。
我查看了之前的issues,看到您回复由于weibo数据规模小,所以存在不稳定的问题。因此我又尝试了8个不同的random…
-
Hi!
First of all, thanks for this great repository. It is well documented and extremely useful to investigate the state-of-the-art in end-to-end / seq2seq models for ASR.
I am trying to apply the …
-
Below is the exec info:
````
batched-wav-nnet3-cuda --acoustic-scale=1.0 --add-pitch=true --beam=15.0 --feature-type=mfcc --frame-subsampling-factor=3 --frames-per-
chunk=150 --gpu-feature-extract=…