Mahesh Bhosale

I see that log_softmax is calculated in forward() of output layer used by all the models. Should we be using BCELoss or CrossEntopyloss in that case?

Issues with custom dataset.

12

I have been trying to do pairwise registration on the CHAOS dataset. https://chaos.grand-challenge.org/Data/ First I converted the volumes from .dcm to .nii. I converted the segmentation labels from .png to...

Understanding the inputs to the network

1

I also referred the blog you mentioned time_sequence_prediction on pytorch forums. I am using a similar approach for one my problem where there are no features but data itself. I...

Required preprocessing for custom dataset?

5

I am trying to use VoxelMorph to register the images in the Liver dataset - [CHAOS](https://chaos.grand-challenge.org/Data/) I have had issues with getting good results on this dataset, in particular, although...

Where to find the weights for cin256_v2/unet.ckpt

1

I am a little confused about how the model weights are organized? I saw in one of the derived works that "models/ldm/cin256-v2/unet.ckpt" path is used for Unet weights and I...

How to split the data into train and test?

1

Hi, I downloaded the google drive data from the repository you mentioned. How to split this data into train and test so as to reproduce your results. I am reluctant...

Any possible steps to do this please? I do not want to install Baidu related softwares on my laptop. I downloaded the data from google drive.

How to use the tokenizer?

1

Thanks for releasing the model on Huggingface. I wanted to use the text encoder. For that I need to tokenize the input. But how to use the tokenizer? Can we...

Custom batch selection for logging

3

### Description & Motivation Need to be able to select the same batch in every logging cycle. For generation pipelines similar to stable diffusion it is very hard to gauge...