-
@vickyFox Hello,good work! And i have a small question.
You mentioned in the paper you directly apply the released pretrained model of CA, EC, PIC, but to my knowledge, EC didn't supply pretrained mo…
cmyyy updated
5 years ago
-
Hello, where do I find the celeba-hq-512-4.8k dataset?
-
The command I executed was
'python main.py --ni --simplified --config celeba_hq.yml --path_y LR --eta 0.85 --deg "sr_bicubic" --deg_scale 1 --sigma_y 0.1 -i LR02', and it resulted in an error showin…
-
Hi. Thank you for sharing!
I am trying to run with :
--ni --config celeba_hq.yml --doc celeba --timesteps 20 --eta 0.85 --etaB 1 --deg sr4 --sigma_0 0.05 -i celeba_hq_sr4_sigma_0.05
but I am gettin…
-
I'm quite confused about FID of CelebA-HQ.
In NCP-VAE and VAEBM paper, it is reported as 40.26, while recent paper LSGM reported as 29.76.
Were there further improvement of NVAE after publicatio…
-
Hello, what a nice work! In your paper, your model uses the Places2 and CelebA-HQ datasets, but these datasets do not have corresponding text information. Can you provide these data and your pre-train…
-
@podgorskiy
Hi, Thanks for the great paper.
I wonder what is the aim of **prepare_celeba_hq_tfrecords.py** comparing to **prepare_celeba_tfrecords.py**.
I have successfully generated celeba…
-
I was trying to run this with docker, it didn't work:
After clone:
$ docker build -t celeba_hq . && docker run -it -v $(pwd):/data celeba_hq
...
PackagesNotFoundError: The following packages are…
-
hey there,
We tested LGNet with CelebA-HQ. According to paper, we selected 3000 pictures randomly from CelebA-HQ and used irregular mask for testing.
It may be that our test config di…
-
Hi! Thank you so much for creating this fantastic project!! It look absolutely amazing.
I tried to download the pre-trained models but all the Yandex links are dead. Do you happen to have the files…