search

StevenLiuWen / ano_pred_cvpr2018

Official implementation of Paper Future Frame Prediction for Anomaly Detection -- A New Baseline, CVPR 2018
435 stars 126 forks source link

Training on new dataset #23

Closed tahmidzbr closed 5 years ago

tahmidzbr commented 5 years ago

Hi, Many thanks for your great work! Can you please help with code if i want to train your algorithm on my own dataset?

tahmidzbr

MillayWang commented 5 years ago

It will help if you look at what the original dataset file contains. You have to add some items in hyper_params.ini and evaluate.py to make your dataset have those parameters like original datasets. You may also need to label ground truth (.mat file) of abnormal behaviors of your own data during evaluating the auc.