The benchmark for AlexNet is part of the Tensorflow repository when I clone it from git. The benchmark implements the layers, but it looks to me that the actual weights for AlexNet are not loaded at any point.
I would like to play around with Tensorflow, but my applications (in caffe) are using a pre-trained AlexNet.
Do you think they will release the weights as well?
I don't know of an existing one, but someone wrote a converter to import Caffe models into tensorflow, and you can find pre-trained Alexnet models for Caffe (also see the BVLC Model-Zoo). I can't promise it will work, but you can quite possibly glue those two together to get what you want.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With