-
Hi,thank you for your work.
Does the input size and output size support other sizes, such as 640*512 or 512*512? If possible, where the code needs to be modified?
Looking forward to your answer.
-
I am trying to get human → anime conversion to run. My current code is as follows:
```python
import argparse
from pathlib import Path
import numpy as np
import skimage.io
from tensorpack imp…
dniku updated
4 years ago
-
Hello, I appreciate your work. Could you please provide `ema_lsun.ckpt`? I don't know where I can download it. Thanks!
-
Something to consider as a highly speculative research project once Ganbooru is done.
---
A distinct trend in recent DL has been self-attention moving beyond sequence or text data to image data …
gwern updated
3 years ago
-
What is the timetable for the code being released?
-
- https://arxiv.org/abs/2004.08697
- 2020
学習の離散化は、観測データの複数の説明因子と生成因子からなる低次元表現を見つけることを目的としている。
観測データから独立した因子を分離するためには、一般的に変分自動符号化器(VAE)の枠組みが用いられる。
しかし、実際のシナリオでは、意味を持つ要因は必ずしも独立ではありません。
むしろ、これらの要…
e4exp updated
2 years ago
-
When I run the command:`python train_ddgan.py --dataset celeba_256 --image_size 256 --exp ddgan_celebahq_exp1 --num_channels 3 --num_channels_dae 64 --ch_mult 1 1 2 2 4 4 --num_timesteps 2 --num_res_b…
-
Can you / will you upload any pretrained model to compare results, please?
-
Please be able to export images with high resolution.
-
Nathaniel Ruiz et al. dropped https://hyperdreambooth.github.io/.
It combines DreamBooth, LoRA, and hypernetworks for a 100kb residual, generated in 20 seconds, that rivals original DreamBooth in …