improved-diffusion
improved-diffusion copied to clipboard
openai
Release for Improved Denoising Diffusion Probabilistic Models
Hi, Thanks for sharing the code. Just wondering how to set our hyper parameters in the flags? I changed some of the hyper parameters in "def model_and_diffusion_defaults():" function in "script_utils.py"....
Hello, I am trying to train the diffusion model on cifar-10 dataset but as mentioned on the readme, i need to setup the hyperparameters, where can i found the? or...
Is it through running "image_sample".py to see the corresponding results?
I noticed a scaling factor of 0.5 when sampling from p(xt-1|xt). I tried to find the definition of this 0.5 on the paper but failed. Is there any special factor...
Thanks for your great work! I am new to MPI and I ran into some nccl errors when I use your command to launch a training. My environment is >...
I trained a model on cifar10 with below settings. `MODEL_FLAGS="--image_size 32 --num_channels 128 --num_res_blocks 3 --learn_sigma True --dropout 0.1" DIFFUSION_FLAGS="--diffusion_steps 4000 --noise_schedule linear" TRAIN_FLAGS="--lr 1e-4 --batch_size 16"` but it overfits...
https://github.com/openai/improved-diffusion/blob/783b6740edb79fdb7d063250db2c51cc9545dcd1/improved_diffusion/unet.py#L192 Do you think that setting the "affine" transform to False in the out_norm layer (which is a a GroupNorm module) would cause any degradation? I am a little bit...
Hi I am training the improved-diffusion model on the CUB dataset to obtain a .ckpt file for the DiffusionCLIP model. Does anybody know how long the training will take? Thanks
Hello, I am using the repo to produce synthetic histopathology images. However, I am observing a weird color shift during training. As training progress, this shift is getting worst and...
Wether I'm using mpi or not, if I have 2 gpus with one node, it will always ignore the second gpu. I am unable to use multiple gpus. How do...
