Mario Larcher

Sagemaker arbitrarily stops copying checkpoints to S3

11

**Describe the bug** Sagemaker arbitrarily stops copying checkpoints to S3, copy those up to epoch 9 (see logs below) and then that's it (although the logs show that more have...

Consider Making Normalization Optional in MappingFeedForwardBlock

2

Hi there! I've noticed that in the `forward` method of `MappingNetwork`, you apply `RMSNorm` to the input: ```python class MappingNetwork(nn.Module): def __init__(self, n_layers, d_model, d_ff, dropout=0.0): super().__init__() self.in_norm = RMSNorm(d_model)...

AttributeError in Oxford Flowers Demo: 'dict' object has no attribute 'convert'

1

Hi, First, congratulations on this amazing repository; it's a great codebase for study. I'm attempting to run the Oxford Flowers demo but am encountering an error. I'm not sure if...

Hi! I think a typo fixed in notebook 01 wasn't fixed in notebook 02, q(xt | x0) ~ N(sqrt(1-alpha)*x0, sqrt(1-alpha)*I) should be q(xt | x0) ~ N(sqrt(1-alpha)*x0, 1-alpha\*I). For the...

In resnet.py there are (more than) two import *: ``` from .conv import * from .init import * ``` Both contain a `conv` def. It took me a moment to...

In notebook 21, the `WandBCB` assumes the presence of a `train` attribute in `_log`, but this is not present. A possible fix would be to replace it with `d["train"] ==...

In ch7 if I execute triangulation `./build/triangulation 1.png 2.png` I get:  Is this correct? With these values less than or close to 10, all key points have the same...