Is it working?

[mwmw7]

I want to train from scratch , but is it working ?

[CreamyLong]

Sure, what task are you trying to train from scratch?

[mwmw7]

Sure, what task are you trying to train from scratch?

Our company (https://toonsquare.co/) really wants to create our own foundation model (text to image, img2img, etc.), similar to Stable Diffusion and DALL·E, because we have tons of webtoon images and descriptions

[CreamyLong]

I suggest that you can have a look of readme to prepare the data and config Then, run the training scripts CUDA_VISIBLE_DEVICES=<GPU_ID> python main.py --base configs/latent-diffusion/<config_spec>.yaml -t --gpus 0,

[mwmw7]

I suggest that you can have a look of readme to prepare the data and config Then, run the training scripts CUDA_VISIBLE_DEVICES=<GPU_ID> python main.py --base configs/latent-diffusion/<config_spec>.yaml -t --gpus 0,

Thank you for your reply

but I got these error message, I think there is not "'cls_txt'" in ldm/data/ path

`Merged modelckpt-cfg: {'target': 'pytorch_lightning.callbacks.ModelCheckpoint', 'params': {'dirpath': 'logs/2024-05-28T20-12-54_txt2img-sdv1/checkpoints', 'filename': '{epoch:06}', 'verbose': True, 'save_last': True, 'monitor': 'val/loss_simple_ema', 'save_top_k': 3}} GPU available: True, used: True TPU available: False, using: 0 TPU cores IPU available: False, using: 0 IPUs

Traceback (most recent call last): File "main.py", line 701, in data.prepare_data() File "/home/mwmw7/anaconda3/envs/ldm/lib/python3.8/site-packages/pytorch_lightning/core/datamodule.py", line 428, in wrapped_fn fn(*args, kwargs) File "/home/mwmw7/PycharmProjects/toonsquare-creamylong-stable-diffusion/main.py", line 197, in prepare_data instantiate_from_config(data_cfg) File "/home/mwmw7/PycharmProjects/toonsquare-creamylong-stable-diffusion/ldm/util.py", line 129, in instantiate_from_config return get_obj_from_str(config["target"])(config.get("params", dict())) File "/home/mwmw7/PycharmProjects/toonsquare-creamylong-stable-diffusion/ldm/util.py", line 137, in get_obj_from_str return getattr(importlib.import_module(module, package=None), cls) File "/home/mwmw7/anaconda3/envs/ldm/lib/python3.8/importlib/init.py", line 127, in import_module return _bootstrap._gcd_import(name[level:], package, level) File "", line 1014, in _gcd_import File "", line 991, in _find_and_load File "", line 973, in _find_and_load_unlocked ModuleNotFoundError: No module named 'ldm.data.cls_txt' `

[mwmw7]

I suggest that you can have a look of readme to prepare the data and config Then, run the training scripts CUDA_VISIBLE_DEVICES=<GPU_ID> python main.py --base configs/latent-diffusion/<config_spec>.yaml -t --gpus 0,

Thank you for your reply

but I got these error message, I think there is not "'cls_txt'" in ldm/data/ path

`Merged modelckpt-cfg:

{'target': 'pytorch_lightning.callbacks.ModelCheckpoint', 'params': {'dirpath': 'logs/2024-05-28T20-12-54_txt2img-sdv1/checkpoints', 'filename': '{epoch:06}', 'verbose': True, 'save_last': True, 'monitor': 'val/loss_simple_ema', 'save_top_k': 3}}

GPU available: True, used: True

TPU available: False, using: 0 TPU cores

IPU available: False, using: 0 IPUs

Traceback (most recent call last):

File "main.py", line 701, in

data.prepare_data()

File "/home/mwmw7/anaconda3/envs/ldm/lib/python3.8/site-packages/pytorch_lightning/core/datamodule.py", line 428, in wrapped_fn

fn(*args, **kwargs)

File "/home/mwmw7/PycharmProjects/toonsquare-creamylong-stable-diffusion/main.py", line 197, in prepare_data

instantiate_from_config(data_cfg)

File "/home/mwmw7/PycharmProjects/toonsquare-creamylong-stable-diffusion/ldm/util.py", line 129, in instantiate_from_config

return get_obj_from_str(config["target"])(**config.get("params", dict()))

File "/home/mwmw7/PycharmProjects/toonsquare-creamylong-stable-diffusion/ldm/util.py", line 137, in get_obj_from_str

return getattr(importlib.import_module(module, package=None), cls)

File "/home/mwmw7/anaconda3/envs/ldm/lib/python3.8/importlib/init.py", line 127, in import_module

return _bootstrap._gcd_import(name[level:], package, level)

File "", line 1014, in _gcd_import

File "", line 991, in _find_and_load

File "", line 973, in _find_and_load_unlocked

ModuleNotFoundError: No module named 'ldm.data.cls_txt'

`

I fixed it

I just rename from "cls_txt" to "custom" I think it needs to be fixed