Loading pre-trained resnet model to cleverhans model format

Question

I am trying to load a pre-trained ResNet model from the MadryLab CIFAR-10 challenge into CleverHans to compute transfer attacks.

However restoring the saved models into the model_zoo.madry_lab_challenges.cifar10_model.ResNet object does not work. It appears to restore fine initially, but when I try to actually use the model, I get an error such as:

Attempting to use uninitialized value ResNet/unit_3_1/residual_only_activation/BatchNorm/moving_mean

The easiest way to reproduce this error is to actually just run the provided attack_model.py example included in CleverHans here:

https://github.com/tensorflow/cleverhans/blob/master/examples/madry_lab_challenges/cifar10/attack_model.py

It encounters the same error after loading the model when it tries to use it, on both adv_trained and naturally_trained models.

Is there a workaround to this problem?

It seems the other option is to use the cleverhans.model.CallableModelWrapper instead, but I haven't been able to find an example of how to use that.

Which version of python, tensorflow, cleverhans are you using? — Nicolas Papernot, Jul 31 '19 at 17:49
@NicolasPapernot python 3.6.8, tensorflow 1.13.1, cleverhans 3.0.1 — quarkral, Aug 03 '19 at 02:11
Would you mind trying with python 2 and an older version of tensorflow? I think the specific example you are pointing to hasn't been updated in a while. My guess would be tensorflow v.1.7 or 1.8 had a different API — Nicolas Papernot, Aug 04 '19 at 15:23

Loading pre-trained resnet model to cleverhans model format

0 Answers0