T2I-Adapter
T2I-Adapter copied to clipboard
TencentARC
T2I-Adapter
When I montage them,I just get this image.Do you make the same experiments?   
I found that T2I-Adapter is trained with LAION 6+ and Lexica. However, I did not find descriptions of the latter. Is this a private dataset? https://github.com/TencentARC/T2I-Adapter/blob/c408b059c36e3f9ce336b66746bd606edaa5483a/configs/train/Adapter-XL-sketch.yaml#L15
 I don't know why, but I followed someone else's video and there's no problem with what's in their video
HI, the paper reports the model is trained with a batch size of 8 on 32G V100, but I got out of memory with the default settings in train_seg.py. When...
Hello, https://github.com/TencentARC/T2I-Adapter/blob/16bba674b472121d5a86e3ed6b935f91d516bc74/train_sketch.py#L231 How do you obtain the mask images of train2017_color? Are you using stuff_train2017_pixelmaps? Look forward for your reply.
Hello ChongMou, May I ask why you use “accelerate launch train_sketch.py --pretrained_model_name_or_path stabilityai/stable-diffusion-xl-base-1.0 --output_dir experiments/adapter_sketch_xl --config configs/train/Adapter-XL-sketch.yaml --mixed_precision="fp16" --resolution=1024 --learning_rate=1e-5 --max_train_steps=60000 --train_batch_size=1 --gradient_accumulation_steps=4 --report_to="wandb" --seed=42 --num_train_epochs 100” for training and...
Hi, Thanks for the nice work! But during the evaluation of your provided model on hugging face, I found that I could not reproduce your provided FID and CLIP score....
What is the purpose of performing * 2-1 operations on images during the first stage of encoding during training？This differs from standard SD training, wouldn't it lead to poor encoding...
