Image
Image
Image
Image
Image
Image
Image
Image
Image
Image



Search
»

LISOR

Description: The package includes the Python code of the LISOR (Learning with Interpretable Structure frOm gated Rnn). You will find a main process 'main.py' in which the models of task '0110' and task '000' are trained. The directory 'data' includes 2 synthetic datasets which are for task '0110' and task '000'. The trained model will be saved in the form of pickle and run 'test.py' will give the results of task '0110' and '000'. The file 'FSA_lib.py' is the supporting file which includes several needed classes and functions.

References:
[1] B.-J. Hou, and Z.-H. Zhou. Learning with Interpretable Structure from Gated RNN. IEEE Transactions on Neural Networks and Learning Systems, 2020, 31(7): 2267-2279.

ATTN: This package only provides the training and testing programs for the task 0110 and task 000. The programs for the real task are not provided due to large scale of the 'word2vec' dataset. However, the inner logic is the same. The researchers could make it in their own ways.

ATTN2: This package is free for academic usage. You can run it at your own risk. For other purposes, please contact Prof. Zhi-Hua Zhou (zhouzh@nju.edu.cn).

ATTN3: This package was developed by Dr. Bo-Jian Hou (houbj@lamda.nju.edu.cn). For any problem concerning the code, please feel free to contact Dr. Hou.

Download: [code] (120KB)
  Name Size

Image
PoweredBy © LAMDA, 2022