What makes HanLP different than the majority of OSS projects?
One of the most important factors would be the large scale professional corpora, and the correct way to make use of them.
To have some unique pretrained LM before releasing the beta version would be a cool idea. Don't you think so?

Description

There is a readme for the repo metrics subfolder under tools. It needs general review to make sure it is accurate.

Other Comments

Principles of NLP Documentation
Each landing page at the folder level should have a ReadMe which explains -
○ Summary of what this folder offers.
○ Why and how it benefits users
○ As applicable - Documentation of using it, brief d

Problem description

Short description

In certain conditions some CRF tags transitions can by missing after the data augmentation or can be "underrepresented".
We must ensure that all possible tags transitions are in the augmented dataset so that inference does not fail systematically on those examples

Example

Given a dataset with 1 intent and 3 slots: slot_1, slot_2, `slot

When passing empty string into predict function, this line and this line will cause the IndexError.

If possible I think it could be changed to if text == "" or text[-1] != '\n': and the problem should be solved.

we use this code to build our project, but we found the acc dropped. So , we review the code, and find the following issues.

1. This code did not implemented "mask" in the "AttLayer" class.
2. we believe "Dense layer" should implemented in the class "AttLayer", instead of using "Dense" out of the class
3. lost "Activation function" in the Dense layer

We made the above changes，and the acc

This is also on page 356.

from nltk.corpus import sentiwordnet as swn

good = swn.senti_synsets('good', 'n')[0]
Traceback (most recent call last):
File "", line 1, in
TypeError: 'filter' object is not subscriptable

读了你的源码，关于这些矩阵names = ['x','y','tx','ty','allx','ally','adj']分别代表什么？
比如allx => the feature vectors of both labeled and unlabeled training docs/words，你的实验数据不都是有标签的嘛，为什么会有unlabeled training docs？
你在论文中说你的节点初始化为one-hot向量，而我在代码中看到你用word嵌入的平均作为doc嵌入输入，这是为什么？the one-hot labels of the labeled training docs又代表什么？
关于这些x,y,tx,ty等等，我比较难懂，请求您抽出时间为我解答，非常感谢

We should revisit our alpha and beta default values. 1.0 is way to large.

(tensorflow) F:\Postgraduate\KaggleLearning\multi-class-text-classification-cnn-rnn-master\multi-class-text-classification-cnn-rnn-master>python predict.py ./t
rained_results_1541818386/ ./data2/samples.csv
D:\Anaconda\anaconda\envs\tensorflow\lib\site-packages\gensim\utils.py:1212: UserWarning: detected Windows; aliasing chunkize to chunkize_serial
warnings.warn("detected Windows; aliasing c

mrr has been implemented as a class in sklearn called "label_ranking_average_precision_score".

https://scikit-learn.org/stable/modules/generated/sklearn.metrics.label_ranking_average_precision_score.html

The code in "Text Classification with Logistic Regression.ipynb" file will be much shorter if it is used.

Hi, would it be possible for the authors to add the accuracy results that they're getting to the README? Right now I'm seeing numbers which are mid 80's for certain models when I know certain people have reported 89/90 with Deep Learning methods on IMDB.