improved-diffusion

improved-diffusion copied to clipboard

question about Vb loss

1

I want to perform an "Upsampling" task, so I choose the low-resolution image as the condition. While the problem is that, if I set learn_sigma = True, the _vb_terms_bpd will...

ljw919

System: Ubuntu 18.04.6 I follow the instruction following to install openmpi: wget https://download.open-mpi.org/release/open-mpi/v5.0/openmpi-5.0.1.tar.gz tar -zxvf openmpi-5.0.1.tar.gz cd openmpi-5.0.1 ./configure --prefix=$HOME/openmpi CC=gcc CXX=g++ --disable-mpi-fortran --disable-mca-dso make make install And I add...

Zhixi-Li

No module named 'improved_diffusion'

7

请问遇到ModuleNotFoundError: No module named 'improved_diffusion' ​该怎样处理呢

liuyisu

A problem about the weight λ of Lvlb

4

In the paper, λ is 0.001. The code sets learn_sigma as True and rescale_learned_sigmas as False, so the loss type will be gd.LossType.MSE, in this loss type ,the Lvlb will...

yinguanchun

How do I produce images of larger size?

rick-dn

What's the meaning of "For fixedlarge, we set initial (log-)variance like os to get a better decoder log likelihood"?

1

Hi, I notice when you define the **model_variance** [here](https://github.com/openai/improved-diffusion/blob/783b6740edb79fdb7d063250db2c51cc9545dcd1/improved_diffusion/gaussian_diffusion.py#L279), first item of variance is set as **posterior_variance[1].** The comment here explains that _" we set initial (log-)variance like so to...

haohang96

minimal number of samples for stable training -> some issue

2

Hi ! I was wondering if some people tested the diffusion training process with few images ( 1000 or less) and obtained good results ? I am working with rare...

NicolasNerr

Problem with training one's own dataset

2

What should we do if this problem occurs ModuleNotFoundError: No module named 'improved_diffusion'

liuyisu

where change my checkpoint dir to save .pt

2

moicesong

Query: Distributed learning

6

I am trying to implement the improved-ddpm project using google colab (1 GPU), I am not sure how to rectify the distributed training problem coming because of the following code...

KomputerMaster64