Default color space transformations

Hi, I was following the CAM chapter of fastai2 tutorials.

path = untar_data(URLs.PETS)/'images'
def is_cat(x): return x[0].isupper()
dls = ImageDataLoaders.from_name_func(
    path, get_image_files(path), valid_pct=0.2, seed=21,
    label_func=is_cat, item_tfms=Resize(224))
learn = cnn_learner(dls, resnet34, metrics=error_rate)
learn.fine_tune(1)

class Hook():
    def hook_func(self, m, i, o): self.stored = o.detach().clone()
hook_output = Hook()
hook = learn.model[0].register_forward_hook(hook_output.hook_func)

img = dls.train.one_batch()[0][0]
x, = first(dls.test_dl([img]))

output = learn.model.eval()(x)
F.softmax(output, dim=-1)

hook_output.stored[0].shape

act = hook_output.stored[0]
cam_map = torch.einsum('ck,kij->cij', learn.model[1][-1].weight, act)

x_dec = TensorImage(dls.train.decode((x,2))[0][0])
_,ax = plt.subplots()
x_dec.show(ctx=ax)
x_dec

I am very confused about the color space here. So I also plotted x and img:

image

I know I loaded the picture from dls. How should I properly display the photo tho? Would greatly appreciate your help. This isn’t the first time that this issue has occurred.