search

abhijitjadhav1998 / Deepfake_detection_using_deep_learning

This projects aims in detection of video deepfakes using deep learning techniques like RestNext and LSTM. We have achived deepfake detection by using transfer learning where the pretrained RestNext CNN is used to obtain a feature vector, further the LSTM layer is trained using the features. For more details follow the documentaion.
GNU General Public License v3.0
495 stars 164 forks source link

Enabling project work on Non-cuda devices #15

Closed Tharun-PS closed 3 months ago

Tharun-PS commented 1 year ago

In the readme file, under "We welcome Open Source Contribution", Enabling working of project on Non Cuda Computers. i.e on normal or AMD GPUs is considered not possible as dlib uses CUDA internally. I have modified the code in such a way that the project works very fine on Normal computer's CPU without requiring GPU. Wishing to contribute to the requirements stated in readme file. I'm using Lenovo Ideapad S340 with intel i5 processor.

abhijitjadhav1998 commented 1 year ago

Thanks, Changes look good, but if you can refactor a bit to provide options for selecting CPU or GPU. Also please don't commit any binary files

Tharun-PS commented 1 year ago

Okay sure. I'll add that and update

Tharun-PS commented 1 year ago

I have made some changes. First it will check if cuda is availble or not. If cuda is available --> device = 'cuda' else --> device = 'cpu' and this device variable will used in code that generalizes the usage for all users.

abhijitjadhav1998 commented 1 year ago

Changes looks good .. but you can try deleting unwanted files like py-cache and all

Tharun-PS commented 1 year ago

Changes looks good .. but you can try deleting unwanted files like py-cache and all

And yes. I have deleted that directory. Sorry for the delay

abhijitjadhav1998 commented 1 year ago

Sorry for the delay, was a bit busy. I can still see some pycache files

IMMORTALCHIRU commented 5 months ago

I have made some changes. First it will check if cuda is availble or not. If cuda is available --> device = 'cuda' else --> device = 'cpu' and this device variable will used in code that generalizes the usage for all users.

how to check the cuda availability??

abhijitjadhav1998 commented 3 months ago

No comments on the PR. Closing the PR