search

khalooei / ALOCC-CVPR2018

Adversarially Learned One-Class Classifier for Novelty Detection (ALOCC)
MIT License
222 stars 76 forks source link

Something about the test.py file #21

Closed liulonghoi closed 4 years ago

liulonghoi commented 4 years ago

Hello, I change the dataset for my own dataset, and the test.py file run well, but i dont kown how this file work, could you tell me what does this file do? How to use a trained model? Thanks in advance!

codeman008 commented 4 years ago

Hello, I am also doing single-type image detection recently. How do you train your data?Thank you very much

khalooei commented 4 years ago

Dear @liulonghoi and @codeman008 , I'm so sorry for late reply because I'm busy right now.

Hello, I change the dataset for my own dataset, and the test.py file run well, but i dont kown how this file work, could you tell me what does this file do? How to use a trained model? Thanks in advance!

Hello, I am also doing single-type image detection recently. How do you train your data?Thank you very much

To use our approach in new dataset, you can just change the modification which related to your dataset. You must check your normalization step, you must check your inputs and etc. Therefore, you can trace the entrance of your model configuration and check each section in training and test phase. You can customize it in different dataset and settings. This work is used in some industrial company in real data in previous years.