Function std::tensor_group_norm

&FloatTensor tensor_group_norm(
    const &FloatTensor x,
    const &FloatTensor gamma,
    const &FloatTensor beta,
    unsigned long channels,
    unsigned long plane,
    unsigned long groups,
    float eps,
)

Group normalization (Wu & He 2018): each image's channels split into 'groups' groups, each normalized over its channels and planes, then gamma / beta per channel. Batch-independent, the same in training and inference. groups 1 is layer norm over (C, H, W); groups == channels is instance norm.