hello,
i am unable to compute GraphiteBlackBox attack on gpu.
My implementation:
from art.estimators.classification import PyTorchClassifier
from art.attacks.evasion import GRAPHITEBlackbox

classifier = PyTorchClassifier(model=model, device_type="cuda",loss=nn.CrossEntropyLoss(), input_shape= ([10,3,112, 112]), nb_classes=8)

noise_size = (112,112)
net_size = (112,112)
attack = GRAPHITEBlackbox(classifier=classifier, noise_size=noise_size, net_size=net_size,num_xforms_mask=50, num_xforms_boost=50, num_boost_queries=500)

image_adv = attack.generate(x=x_test_class_0, y=y_test_class_0, x_tar=x_tar_replicated)

Where
x=x_test_class_0, y=y_test_class_0, x_tar=x_tar_replicated are all pytorch tensors

which generates following error:

TypeError Traceback (most recent call last)
in <cell line: 18>()
16
17
---> 18 image_adv = attack.generate(x=x_test_class_0, y=y_test_class_0, x_tar=x_tar_replicated)

2 frames
/usr/local/lib/python3.10/dist-packages/art/attacks/evasion/graphite/graphite_blackbox.py in generate(self, x, y, **kwargs)
241 clip_min, clip_max = self.estimator.clip_values
242 else:
--> 243 clip_min, clip_max = np.min(x), np.max(x)
244
245 # target initialization image

/usr/local/lib/python3.10/dist-packages/numpy/core/fromnumeric.py in min(a, axis, out, keepdims, initial, where)
2951 6
2952 """
-> 2953 return _wrapreduction(a, np.minimum, 'min', axis, None, out,
2954 keepdims=keepdims, initial=initial, where=where)
2955

/usr/local/lib/python3.10/dist-packages/numpy/core/fromnumeric.py in _wrapreduction(obj, ufunc, method, axis, dtype, out, **kwargs)
84 return reduction(axis=axis, dtype=dtype, out=out, **passkwargs)
85 else:
---> 86 return reduction(axis=axis, out=out, **passkwargs)
87
88 return ufunc.reduce(obj, axis, dtype, out, **passkwargs)

TypeError: min() received an invalid combination of arguments - got (out=NoneType, axis=NoneType, ), but expected one of:

• ()
• (Tensor other)
• (int dim, bool keepdim)
  didn't match because some of the keywords were incorrect: out, axis
• (name dim, bool keepdim)
  didn't match because some of the keywords were incorrect: out, axis

Any help is appreciated.