search

seominseok0429 / Real-world-Anomaly-Detection-in-Surveillance-Videos-pytorch

Real-world Anomaly Detection in Surveillance Videos- pytorch Re-implementation
102 stars 24 forks source link

Lossing models #18

Open zimengxueying opened 1 year ago

zimengxueying commented 1 year ago

Thanks for your great work! You have submit the full code by the https://drive.google.com/file/d/1xYsBiCSmXjE0BwoiH_Bcm4AWaA1pESHF/view.However,when I run the vis.py, it's wrongModuleNotFoundError: No module named 'models.model'I think the model directory doesn't appear the model and can not impot generater.So,would you please help me? thx so much

suki1504 commented 1 year ago

@zimengxueying

please check this link https://drive.google.com/file/d/1xYsBiCSmXjE0BwoiH_Bcm4AWaA1pESHF/view

zimengxueying commented 1 year ago

@zimengxueying

please check this link https://drive.google.com/file/d/1xYsBiCSmXjE0BwoiH_Bcm4AWaA1pESHF/view

Thank you so much! I've found it !

suki1504 commented 1 year ago

@zimengxueyin @seominseok0429

Are you working on this code ??? If yes I have some doubts can you please clarify that

After successful completion of training on how to test it with the trained model, after training what can I do to test on video

zimengxueying commented 1 year ago

@zimengxueyin @seominseok0429

Are you working on this code ??? If yes I have some doubts can you please clarify that

After successful completion of training on how to test it with the trained model, after training what can I do to test on video

I have used this repo before the full code uploaded. But I think if you want to test the model to get AUC you can just run main.py. And if you want to use the trained model ,I think you can use the codes in directory VIS. According to the file name, I guess video2frame.py should run first to spilit the video, and use vis.py to detect whether the frame is abnormal. You could see the code for more details.

marcocapp commented 1 year ago

@zimengxueyin @seominseok0429 Are you working on this code ??? If yes I have some doubts can you please clarify that After successful completion of training on how to test it with the trained model, after training what can I do to test on video

I have used this repo before the full code uploaded. But I think if you want to test the model to get AUC you can just run main.py. And if you want to use the trained model ,I think you can use the codes in directory VIS. According to the file name, I guess video2frame.py should run first to spilit the video, and use vis.py to detect whether the frame is abnormal. You could see the code for more details.

The checkpoint in the Vis directory is different from the one you can generate from the main with default settings, do you know why?

zimengxueying commented 1 year ago

@zimengxueyin @seominseok0429 Are you working on this code ??? If yes I have some doubts can you please clarify that After successful completion of training on how to test it with the trained model, after training what can I do to test on video

I have used this repo before the full code uploaded. But I think if you want to test the model to get AUC you can just run main.py. And if you want to use the trained model ,I think you can use the codes in directory VIS. According to the file name, I guess video2frame.py should run first to spilit the video, and use vis.py to detect whether the frame is abnormal. You could see the code for more details.

The checkpoint in the Vis directory is different from the one you can generate from the main with default settings, do you know why?

Sorry..I used the checkpoint he provided, and I didn't test the checkpoint trained from main.py. If you have any idea, could you tell me

marcocapp commented 1 year ago

@zimengxueyin @seominseok0429 Are you working on this code ??? If yes I have some doubts can you please clarify that After successful completion of training on how to test it with the trained model, after training what can I do to test on video

I have used this repo before the full code uploaded. But I think if you want to test the model to get AUC you can just run main.py. And if you want to use the trained model ,I think you can use the codes in directory VIS. According to the file name, I guess video2frame.py should run first to spilit the video, and use vis.py to detect whether the frame is abnormal. You could see the code for more details.

The checkpoint in the Vis directory is different from the one you can generate from the main with default settings, do you know why?

Sorry..I used the checkpoint he provided, and I didn't test the checkpoint trained from main.py. If you have any idea, could you tell me

Nope I tried everything I guess it was taken from another repository