Comments (15)
q(x_0) stands for the whole data distribution, i.e your training dataset.
We can not explicitly express the correct q(x_0) (we do not know if q(x_0) is Gaussian or other distribution), we can only draw samples from the data distribution q(x_0)
from ddpm-ip.
yes
from ddpm-ip.
Thank you. Also, for example, I’m working on cifar10 dataset. Then the dimension of x0,⋯,xt is 32×32×3, right?
yes
from ddpm-ip.
Thank you! I wonder what you mean by "each FID value is computed using
T = 1000
sampling steps". Does it implydiffusion_steps 1000
in the code below? Thanks.mpirun python scripts/image_sample.py \ --image_size 32 --timestep_respacing 100 \ --model_path PATH_TO_CHECKPOINT \ --num_channels 128 --num_head_channels 32 --num_res_blocks 3 --attention_resolutions 16,8 \ --resblock_updown True --use_new_attention_order True --learn_sigma True --dropout 0.3 \ --diffusion_steps 1000 --noise_schedule cosine --use_scale_shift_norm True --batch_size 256 --num_samples 50000
In Figure 3 of your paper, you calculated FID scores using
T = 1000
sampling steps.
The code uses 100 sampling steps. Figure 3 will be updated in paper later
from ddpm-ip.
Got it. Could you tell me which parameter determines the number of sampling steps in the code below? Thank you.
mpirun python scripts/image_sample.py \ --image_size 32 --timestep_respacing 100 \ --model_path PATH_TO_CHECKPOINT \ --num_channels 128 --num_head_channels 32 --num_res_blocks 3 --attention_resolutions 16,8 \ --resblock_updown True --use_new_attention_order True --learn_sigma True --dropout 0.3 \ --diffusion_steps 1000 --noise_schedule cosine --use_scale_shift_norm True --batch_size 256 --num_samples 50000
--timestep_respacing 100
from ddpm-ip.
Thank you. I'm still a little confused about the notations in the paper. You mentioned in the paper "When training, we always use T=1000 steps for all the models. At inference time, the results reported with T′<T sampling steps have been obtained using the respacing technique." So in here T=1000 refers to
diffusion_steps 1000
, and T′ refers to the parametertimestep_respacing
. Am I right? Thanks.
yes
from ddpm-ip.
I see, thank you! And in
from ddpm-ip.
Thank you. Also, for example, I’m working on cifar10 dataset. Then the dimension of
from ddpm-ip.
Thank you! I wonder what you mean by "each FID value is computed using T = 1000
sampling steps". Does it imply diffusion_steps 1000
in the code below? Thanks.
mpirun python scripts/image_sample.py \
--image_size 32 --timestep_respacing 100 \
--model_path PATH_TO_CHECKPOINT \
--num_channels 128 --num_head_channels 32 --num_res_blocks 3 --attention_resolutions 16,8 \
--resblock_updown True --use_new_attention_order True --learn_sigma True --dropout 0.3 \
--diffusion_steps 1000 --noise_schedule cosine --use_scale_shift_norm True --batch_size 256 --num_samples 50000
In Figure 3 of your paper, you calculated FID scores using T = 1000
sampling steps.
from ddpm-ip.
Thank you! I wonder what you mean by "each FID value is computed using
T = 1000
sampling steps". Does it implydiffusion_steps 1000
in the code below? Thanks.mpirun python scripts/image_sample.py \ --image_size 32 --timestep_respacing 100 \ --model_path PATH_TO_CHECKPOINT \ --num_channels 128 --num_head_channels 32 --num_res_blocks 3 --attention_resolutions 16,8 \ --resblock_updown True --use_new_attention_order True --learn_sigma True --dropout 0.3 \ --diffusion_steps 1000 --noise_schedule cosine --use_scale_shift_norm True --batch_size 256 --num_samples 50000
In Figure 3 of your paper, you calculated FID scores using
T = 1000
sampling steps.
from ddpm-ip.
Got it. Could you tell me which parameter determines the number of sampling steps in the code below? Thank you.
mpirun python scripts/image_sample.py \
--image_size 32 --timestep_respacing 100 \
--model_path PATH_TO_CHECKPOINT \
--num_channels 128 --num_head_channels 32 --num_res_blocks 3 --attention_resolutions 16,8 \
--resblock_updown True --use_new_attention_order True --learn_sigma True --dropout 0.3 \
--diffusion_steps 1000 --noise_schedule cosine --use_scale_shift_norm True --batch_size 256 --num_samples 50000
from ddpm-ip.
Thank you. I'm still a little confused about the notations in the paper. You mentioned in the paper "When training, we always use diffusion_steps 1000
, and timestep_respacing
. Am I right? Thanks.
from ddpm-ip.
Thanks! By the way, I'm trying to train a new model for MNIST dataset using your code and I've created a notebook to download MNIST dataset and transform the training dataset to npz file. The only issue I have is that the dimension of the images in my npz file is 28×28×3, but the default dimension of MNIST images is 28×28×1.
So I wonder if this discrepancy will influence the training of DDPM-IP. Here is the Colab notebook. Thank you.
The default dimension:
from ddpm-ip.
Hi, I also wonder what the number of total_batch_size
is when you were training on the Celeba dataset. I guess total_batch_size
is 8*16=128 since there are two nodes. And how long does the training process take? Thank you.
The code for CelebA 64x64 training:
mpiexec -n 16 python scripts/image_train.py --input_pertub 0.1 \
--data_dir PATH_TO_DATASET \
--image_size 64 --use_fp16 True --num_channels 192 --num_head_channels 64 --num_res_blocks 3 \
--attention_resolutions 32,16,8 --resblock_updown True --use_new_attention_order True \
--learn_sigma True --dropout 0.1 --diffusion_steps 1000 --noise_schedule cosine --use_scale_shift_norm True \
--rescale_learned_sigmas True --schedule_sampler loss-second-moment --lr 1e-4 --batch_size 16
from ddpm-ip.
Hi, I also wonder what the number of
total_batch_size
is when you were training on the Celeba dataset. I guesstotal_batch_size
is 8*16=128 since there are two nodes. And how long does the training process take? Thank you.The code for CelebA 64x64 training:
mpiexec -n 16 python scripts/image_train.py --input_pertub 0.1 \ --data_dir PATH_TO_DATASET \ --image_size 64 --use_fp16 True --num_channels 192 --num_head_channels 64 --num_res_blocks 3 \ --attention_resolutions 32,16,8 --resblock_updown True --use_new_attention_order True \ --learn_sigma True --dropout 0.1 --diffusion_steps 1000 --noise_schedule cosine --use_scale_shift_norm True \ --rescale_learned_sigmas True --schedule_sampler loss-second-moment --lr 1e-4 --batch_size 16
total_batch_size is 8*16=128 is correct. training celeba takes 4-5 days using 16 V100 GPUs
from ddpm-ip.
Related Issues (20)
- A way to create cifar_train.npz HOT 6
- Line 868: noise = th.randn_like(x_start) remains unchanged HOT 5
- train_data_batch in imagenet32 is not npz file HOT 5
- An error when using a different noise schedule HOT 15
- A discrepancy between the table in the paper and the figure in the repo HOT 1
- the pretrained CelebA 64x64 model gives me noisy sample HOT 1
- somebody update 256 x 256 CelebA pretrained with ip?
- Reference batch for calculating FID of ImageNet HOT 1
- FID in Cifar10. HOT 2
- How to train ADM-Baseline.pt on CiFar-10? HOT 6
- CelebA dataset HOT 18
- Something Wrong in imagenet32_npz.py HOT 2
- Noisy Images when training on custom dataset HOT 3
- How to train ADM-IP.pt on CIFAR10 HOT 21
- FID on cifar10 HOT 2
- Input perturbation by increasing the noise strength HOT 3
- FID on CelebA 64x64 HOT 1
- ModuleNotFoundError: No module named 'guided_diffusion' HOT 1
- 你好,我想问加上这个input_pertub之后是否会影响加载之前训练过的checkpoint呢? HOT 1
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 ddpm-ip.