Carlos Nácher Collado

**UPDATE:** **To ensure that even if `infinite=True` we have reproducibility, the second option must be done**. With the first one, if other lines of code calling `np.random` are executed, this...

> Hola, tengo una pregunta relacionada con el uso de pesos de imagenet en imágenes de entrada 3D de 1 canal. ¿Puedo saber qué método se puede usar para adaptar...

Hi! What do you exactly mean? I have my `model.pt` (trained) and also, yes, I have some script that loads the torch model, the data, preprocesses data, does inference, etc....

Hi, have you found any solution? I am struggling with the same. Thank you!

> Hi @samet-akcay! Just to make sure I've gotten this right I need to : > > 1. Create a file named mlflow at src/anomalib/loggers. > 2. That file should...

**Update:** The problem is the number of channels. If I try `input_shape=(32, 96, 96, 1)` instead of `input_shape=(32, 96, 96, 3)` it works. Can't use Unet for single channel volumes?

> Simplest solution is to triplicate input for model) @ZFTurbo This is a possibility, but it limits my batch size to 3 times smaller. If there is no alternative, it...

@ZFTurbo > 1. If you increase input from 1 to 3, the required memory for model almost won't change. If it changes, it is in fact three times as much....

HD95 sorry. It means Hausdorff Distance (percentile 95).[WikiPedia Link](https://en.m.wikipedia.org/wiki/Hausdorff_distance#:~:text=The%20Hausdorff%20distance%20is%20the,point%20in%20the%20other%20set.) In [medpy](https://loli.github.io/medpy/generated/medpy.metric.binary.hd.html#medpy.metric.binary.hd) library it is implemented, but I don’t know how to use as Keras metric.