extra_requirements.json

{
    "numpy": ["core"],
    "requests": ["core"],
    "PyYAML": ["core"],
    "pytest": ["test"],
    "document-utils": ["core"],
    "sphinx-rtd-theme>=0.5.0": ["test"],
    "imageio": ["encoders-image", "encoders-image-tfhub", "clip"],
    "scikit-image":  ["encoders-image", "encoders-image-tfhub", "clip"],
    "soundfile": ["encoders-audio-tfhub"],
    "librosa": ["audio-encoder", "encoders-audio-tfhub"],
    "tensorflow~=2.4.3": ["encoders-text-tfhub", "encoders-audio-tfhub", "encoders-image-tfhub", "encoders-text-tf-transformers", 
        "encoders-text-tfhub-windows", "encoders-image-tf-face-detection"],
    "tensorflow-hub~=0.12.0": ["encoders-text-tfhub", "encoders-audio-tfhub", "encoders-image-tfhub", "encoders-text-tfhub-windows"],
    "tensorflow-text~=2.4.3": ["encoders-text-tfhub"],
    "tf-models-official==2.4.0": ["encoders-text-tfhub", "encoders-text-tfhub-windows"],
    "bert-for-tf2==0.14.9": ["encoders-text-tfhub", "encoders-text-tfhub-windows"],
    "sentence-transformers": ["encoders-text-sentence-transformers"],
    "torch>=1.6.0": ["encoders-audio-pytorch", "encoders-text-torch-transformers", "encoders-text-sentence-transformers", 
        "encoders-image-fastai", "encoders-code-transformers"],
    "fairseq": ["encoders-audio-pytorch"],
    "transformers": ["encoders-text-torch-transformers", "encoders-text-tf-transformers", "encoders-code-transformers"],
    "moviepy": ["encoders-video"],
    "opencv-python": ["encoders-video", "encoders-image-tf-face-detection", "encoders-image-cv2", "clip"],
    "appdirs": ["encoders-image-tf-face-detection"],
    "fastai==2.1.8": ["encoders-image-fastai"],
    "mtcnn": ["encoders-image-tf-face-detection"],
    "Pillow": ["encoders-image-tf-face-detection", "clip"],
    "clip-by-openai": ["clip"],
    "pytorch==1.7.1": ["clip"]
}