Here you can find the model from rcmalli converted to tf.keras.
Original from rcmalli can be found here
Install via:
pip install git+https://github.com/JanderHungrige/tf.keras-vggface
like this:
from keras_vggface_TF.vggfaceTF import VGGFace
from keras_vggface_TF import utils
If pretrained weigths are used, the weigths are automatically downloaded from rcmalli (links included in utils.py).
from keras_vggface_TF.vggfaceTF import VGGFace
pretrained_model = VGGFace(model='resnet50', include_top=True, input_shape=(224, 224, 3), pooling='avg') # pooling: None, avg or max
Tensorflow>=1.15.3
weights from quantisation aware training will be uploaded by me in time.