Jan Eric Lenssen

Hi, I just tried it and it worked for me. The relevant parameter in config is '--data_dir' (it is used for the data_path parameter in utils.py). paths[0] should be 'data/CelebA/splits/train/.jpg'...

Hi, In config.py, the data_format ist chosen based on whether you use GPU or not: ```python if config.use_gpu: data_format = 'NCHW' else: data_format = 'NHWC' ``` I do not know...

I second that. I think it should be dim=0. However, it does not train successfully if i change it.

Thanks for your comments! Using the convolutional features instead of the learned representation is a good hint, although it is not what the authors claim they do. Unfortunately, I do...

Hi, If the module was not found I suspect something went wrong during installation. Can you post the installation log and check if the module appears in the list of...

Hi, The function moves all input capsules (poses + agreements) of one receptive field into an additional depth dimension of the array so that afterwards receptive field aggregation can be...

Hi, We have not run experiments on such large input images yet. The architecture you proposed should run into problems with number of output capsules since you do not reduce...