Skip to content

Latest commit

 

History

History
17 lines (8 loc) · 649 Bytes

README.md

File metadata and controls

17 lines (8 loc) · 649 Bytes

Change Retrieval Systems Implementations Using AutoEncoder Structures

First download the SECOND dataset from the url given below http://www.captain-whu.com/PROJECT/SCD/

Then generate the test and train sets using the create_test_train_set python script

If you want you can change the training seed or edit the train and test ratios with this scripts,

Run second_label_generate script to generate the binary change information for each image in the second dataset.

Optionally Train your model using the train.py function

Test your model performance on the test set, the performance will be given as the precision of the each change label.