Suvojit Manna

I generated the files and stored it as convenient zip files. You can download from Google Drive or generate using pre-built docker image. Visit [suvojit-0x55aa/celebA-HQ-dataset-download](https://github.com/suvojit-0x55aa/celebA-HQ-dataset-download)

@shartoo As quoted in the Progressive Growing of GANs [paper](https://research.nvidia.com/sites/default/files/pubs/2017-10_Progressive-Growing-of/karras2018iclr-paper.pdf) > In this section we describe the process we used to create the high-quality version of the CELEBA dataset, consisting...

@Cuilie you can use [this file](https://raw.githubusercontent.com/nperraud/download-celebA-HQ/master/image_list.txt).

@aciobanusebi as far as I remember that's the memory needed for raw dataset, the deltas, and the generated dataset.

I'm facing the same problem while training from scratch the loss starts at [10,9] the model gives brown images for the initial epoch. So I started finetuning with the provided...

Thanks for this @ffleader1, would you happen to have the accuracy scores too. I'll add it to the README with acknowledgement of course.

Just the script I've put here, Maybe it need some changes due to how training in RX6800 libs are handled ?

Hi, I'm trying to write a [PyTorch port of this](https://github.com/suvojit-0x55aa/3DMMasSTN-Pytorch). Can anyone help me with the code for creation of `BFM_UV.mat` and `prepareExpressionsBFM.m` ?

Any update on this ?

Found the [xavier.m](https://github.com/vedaldi/practical-cnn-reg/blob/master/xavier.m) file in this repo.