Installing the environments

[CauchyComplete]

Hi,

Nice repo!

I'm trying to run a single training code, but setting up the environments has been tricky.
It seems like this repo requires more packages than those listed in requirements.txt.

Could you please update the requirements, or let me know how to install all the necessary dependencies?

Thanks!

[dddb11]

Hi, thanks for your interest! We've noticed this issue too — it's caused by the growing number of models in ForensicHub, which brings in more dependencies. Since users may not need all models, we’ll release a lazy-loading version soon. That way, packages will only be checked when a model is actually used. We also suggest installing any missing packages only when errors pop up — this helps avoid unnecessary bloat.

[CauchyComplete]

@dddb11 Thanks for your reply. Would you mind uploading the yaml file here? You can easily get via

conda env export > environment.yml

OR

pip freeze > requirements.txt

I've installed more than 20 packages I guess, one by one, but it does not end!!

[dddb11]

@dddb11 Thanks for your reply. Would you mind uploading the yaml file here? You can easily get via

conda env export > environment.yml

OR

pip freeze > requirements.txt

I've installed more than 20 packages I guess, one by one, but it does not end!!

Here is the list of dependencies exported using pipreqs, in case it's helpful. Since ForensicHub is compatible with DeepfakeBench (which hasn’t been uploaded to PyPI), you’ll need to clone our forked version Site locally and install it using: pip install -e ..

albumentations==1.3.0 appdirs==1.4.4 colorama==0.4.6 deepfakebench.egg==info efficientnet_pytorch==0.7.1 einops==0.8.1 ftfy==6.3.1 grad_cam==1.5.4 imageio==2.31.6 IMDLBenCo==0.1.41 imdlbenco.egg==info imutils==0.5.4 jpegio==0.2.8 kmeans_pytorch==0.3 lmdb==1.6.2 matplotlib==3.8.2 numpy==2.3.1 opencv_python==4.8.1.78 opencv_python_headless==4.8.1.78 Pillow==10.0.1 Pillow==11.3.0 pytorch_wavelets==1.3.0 regex==2024.11.6 Requests==2.32.4 rich==14.0.0 scikit_learn==1.3.2 segmentation_models_pytorch==0.5.0 setuptools==78.1.0 six==1.16.0 tensorflow==2.19.0 timm==1.0.17 torch==2.0.1 torchvision==0.15.2 tqdm==4.67.1 ttach==0.0.3 yacs==0.1.8