Hi, thank you again for sharing this great work. I used the code for image-to-image translation. When I tested the model, it gave me a different output (completely different from the input). I thought the model was overfitted. I have 11870 folders. I set up the flags as follows:
MODEL_FLAGS="--image_size 128 --num_channels 128 --class_cond False --num_res_blocks 2 --num_heads 1 --learn_sigma True --use_scale_shift_norm False --attention_resolutions 16"
DIFFUSION_FLAGS="--diffusion_steps 1000 --noise_schedule linear --rescale_learned_sigmas False --rescale_timesteps False"
TRAIN_FLAGS="--lr 1e-4 --batch_size 4"
I changed the dropout, learning rate, reduced diffusion steps, and reduced num_channels. I also tried with other datasets. but got the same wrong results. Could you suggest how to solve this problem? and thank you.
Hi, thank you again for sharing this great work. I used the code for image-to-image translation. When I tested the model, it gave me a different output (completely different from the input). I thought the model was overfitted. I have 11870 folders. I set up the flags as follows: MODEL_FLAGS="--image_size 128 --num_channels 128 --class_cond False --num_res_blocks 2 --num_heads 1 --learn_sigma True --use_scale_shift_norm False --attention_resolutions 16" DIFFUSION_FLAGS="--diffusion_steps 1000 --noise_schedule linear --rescale_learned_sigmas False --rescale_timesteps False" TRAIN_FLAGS="--lr 1e-4 --batch_size 4"
I changed the dropout, learning rate, reduced diffusion steps, and reduced num_channels. I also tried with other datasets. but got the same wrong results. Could you suggest how to solve this problem? and thank you.