[Bert multi label text classification practice] I - overview of practical projects

[1] The overview

  【BERT- Multi label text classification practice 】 There are six articles in this series ：

  【BERT- Multi label text classification practice 】 One of —— Overview of practical projects
  【BERT- Multi label text classification practice 】 The second ——BERT The status and explanation of terms
  【BERT- Multi label text classification practice 】 A third —— Data set selection and preprocessing
  【BERT- Multi label text classification practice 】 Of the four ——BERT Selection and of model base Transformers
  【BERT- Multi label text classification practice 】 Of the five —— Data loading
  【BERT- Multi label text classification practice 】 Of the six —— Modeling and training - assessment - test

---

   For now , If the direction is text classification , English text classification 、 Chinese text classification needs to be mastered .

   Many datasets are in English , For example, multi label text dataset 、 Hierarchical label text dataset , If you want to go further in the direction , You have to learn to deal with English texts .

   With the development in recent years ,bert It is one of several models that perform super well , So it's necessary to run bert Related models .

---

[2] Code to get the address

   If you need the source code of this component , Please scan and follow my official account , reply “bert actual combat ”.

   The code structure is shown in the figure ： This project uses pytorch Realization .

[3] Knowledge to be mastered before actual combat

  1、 understand pytorch Basic operation ;

  2、 master Embedded layer 、 Pre training word vector The concept of ;

  3、 master Neural network model Related basic concepts ;

  4、 understand bert Knowledge of models ;

[4] Another project ：【 English single label text classification practice 】

   You can also browse my other project blog ：【 English single label text classification practice 】.

  >= Click here to browse =<