I'm following the examples as close as possible, even using dumpsters in the preparation example script (stable_txt2img.py). But when I'm running the main.py I'm getting an error message:
File "main.py", line 832, in <module>
trainer.fit(model, data)
...
pytorch_lightning.utilities.exceptions.MisconfigurationException: Total length of `Dataloader` across ranks is zero. Please make sure that it returns at least 1 batch.
I assume that the --reg_data_root should be pointing to the files generated from the stable_txt2img.py files and the --data_root is the output dir? I set --class_word dumpster. Not sure what I'm missing.
I assume that data_root is not the output_dir, it's like the training set of other models. You can download some images online and put it in the data_root. I hope it will help you.
I'm following the examples as close as possible, even using dumpsters in the preparation example script (stable_txt2img.py). But when I'm running the main.py I'm getting an error message:
I assume that the --reg_data_root should be pointing to the files generated from the stable_txt2img.py files and the --data_root is the output dir? I set --class_word dumpster. Not sure what I'm missing.
Thanks!