shape error

[16SeungHun]

hi, my environment is 3090ti x 3

i have a shape error

RuntimeError: Given groups=1, weight of size [128, 3, 3, 3], expected input[512, 1, 512, 3] to have 3 channels, but got 1 channels instead

and here is my command

3x3090ti:

BATCH_SIZE = 3 N_GPUS = 3 ACCUMULATE_BATCHES = 1

gpu_list = ",".join((str(x) for x in range(N_GPUS))) + "," print(f"Using GPUs: {gpu_list}")

get_ipython().system('(python main.py -t --base configs/stable-diffusion/pokemon.yaml --gpus "$gpu_list" --scale_lr False --num_nodes 1 --check_val_every_n_epoch 10 --finetune_from /home/shlee/.cache/huggingface/hub/models--CompVis--stable-diffusion-v-1-4-original/snapshots/0834a76f88354683d3f7ef271cadd28f4757a8cc/sd-v1-4-full-ema.ckpt data.params.batch_size= "$BATCH_SIZE" )')

[shub-kris]

I am also getting the same error when running on a single GPU

[16SeungHun]

thx. i try it

[devonbrackbill]

Was anyone able to resolve this? I'm seeing the same issue.

[YIFanH]

I update get_input fuc in ddpm resolve this question as follow: @devonbrackbill x = x[..., None] —> x = x[None,...]， but I meet other question str is not to(device);