You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,
Congratulations for the amazing work. I have some doubts regarding rms normalization.
Which dimensions should be considered for normalization of a CNN?? In the torch code, default axis is -1 which means Width dimension in pytorch CNN. However, in tensorflow it is channels.
Can the normalization be applied on other dimensions as well?? Like in CIFAR-10 experiments. LayerNorm was applied on width and height dimensions.
Thank you.
The text was updated successfully, but these errors were encountered:
The PyTorch (rmsnorm_torch) and TensorFlow (rmsnorm_tensorflow) code do NOT consider the case of CNN. By default, the code can be used for RNN, Feed-Forward and Attention networks, and the normalization is applied to the last dimension.
For the normalization of CNN, I follow the LayerNorm and apply it to the width and height dimensions. Please refer to the CIFAR-10 Classification Section in README for more details.
Hi,
Congratulations for the amazing work. I have some doubts regarding rms normalization.
Which dimensions should be considered for normalization of a CNN?? In the torch code, default axis is -1 which means Width dimension in pytorch CNN. However, in tensorflow it is channels.
Can the normalization be applied on other dimensions as well?? Like in CIFAR-10 experiments. LayerNorm was applied on width and height dimensions.
Thank you.
The text was updated successfully, but these errors were encountered: