Comments (8)
I think the multi scale input may not be produced by resize, it's produced from every to_rbg layer like https://github.com/JiauZhang/GigaGAN/blob/main/model.py#L254
from gigagan-pytorch.
@gooobot ohh got it! so the generator will need to output all the
rgb
at all stages, and be able to pass it into the discriminator. i can get that done tomorrow morning, thank you! 🙏
if a real image taken from the dataset is fed into the discriminator, then it would be resized as the code does now?Yes, the generator will output images with [4x, 8x, ..., 32x, 64x] resolution when use_multi_scale is enabled.
That's what I mean! @lucidrains Thank you for you great effort!
from gigagan-pytorch.
@potato123-hash hello! so i'm actually doing that automatically here. let me know if i misunderstood that part of the paper
from gigagan-pytorch.
@gooobot ohh got it! so the generator will need to output all the rgb
at all stages, and be able to pass it into the discriminator. i can get that done tomorrow morning, thank you! 🙏
if a real image taken from the dataset is fed into the discriminator, then it would be resized as the code does now?
from gigagan-pytorch.
@gooobot ohh got it! so the generator will need to output all the
rgb
at all stages, and be able to pass it into the discriminator. i can get that done tomorrow morning, thank you! 🙏if a real image taken from the dataset is fed into the discriminator, then it would be resized as the code does now?
Yes, the generator will output images with [4x, 8x, ..., 32x, 64x] resolution when use_multi_scale is enabled.
from gigagan-pytorch.
@potato123-hash @gooobot ok, makes sense! will get this all fixed tomorrow! thank you for raising this issue
from gigagan-pytorch.
@potato123-hash do you want to try 0.0.16 and see if that fits your intuition?
import torch
from gigagan_pytorch.gigagan_pytorch import (
TextEncoder,
Generator,
Discriminator,
StyleNetwork
)
text_encoder = TextEncoder(
dim = 512,
depth = 2
).cuda()
discr = Discriminator(
dim = 64,
dim_max = 512,
image_size = 256,
text_encoder = text_encoder,
use_glu = True,
num_skip_layers_excite = 4,
unconditional = False
).cuda()
style_network = StyleNetwork(
dim = 64,
depth = 4,
dim_text_latent = text_encoder.dim
).cuda()
generator = Generator(
dim = 64,
style_network = style_network,
text_encoder = text_encoder,
image_size = 256,
dim_max = 512,
use_glu = True,
num_skip_layers_excite = 4
).cuda()
# mock data
real_images = torch.randn(1, 3, 256, 256).cuda()
texts = ['a happy dog wagging her tail']
# generator
image, rgbs = generator(
texts = texts,
batch_size = 1,
return_all_rgbs = True
)
# discriminator
logits, *_ = discr(
image,
rgbs,
real_images = real_images,
texts = texts
)
from gigagan-pytorch.
feel free to reopen if it isn't resolved!
from gigagan-pytorch.
Related Issues (20)
- Possible Discrepancies HOT 3
- The training code not deal with paired data yet? HOT 2
- [Question] About the upscaler HOT 2
- Multi GPU training HOT 4
- Multi GPU with gradient accumulation
- [Request] Please provide a replicate.com version
- Confused about this project?
- NaN losses after hours of training (UPSAMPLER) HOT 16
- How to implement this model to enhance my input images? Do I have to train the model to use? HOT 2
- Weights of Gigagan Upscaler HOT 1
- Turn on/off gradients computation between generator/discriminator HOT 2
- Wrong order of resolutions list HOT 1
- to_rgb branch has only 1 learnable kernel HOT 7
- Gradient Penalty is very high in the start HOT 10
- How to use this model for SR ?
- Has Anyone Trained This Model Yet? HOT 2
- The text-to-image tasks
- Config to reproduce paper
- question about code in unet_upsampler.py HOT 1
- the loss became nan after a few train steps HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from gigagan-pytorch.