so how does it do it with no captions now? I do see when I do dataset preparations it does add 2 add

G

GomezBroOP•9/3/24, 9:04 AM

thanks

GGomezBro so how does it do it with no captions now? I do see when I do dataset preparatio...

F

Furkan Gözükara SECourses•9/3/24, 9:05 AM

reads folder name

F

Furkan Gözükara SECourses•9/3/24, 9:05 AM

you really should watch video

F

Furkan Gözükara SECourses•9/3/24, 9:05 AM

i tell all of these and more

F

Furkan Gözükara SECourses•9/3/24, 9:05 AM

https://youtu.be/nySGu12Y05k

YouTubeSECourses

FLUX LoRA Training Simplified: From Zero to Hero with Kohya SS GUI ...

Ultimate Kohya GUI FLUX LoRA training tutorial. This tutorial is product of non-stop 9 days research and training. I have trained over 73 FLUX LoRA models and analyzed all to prepare this tutorial video. The research still going on and hopefully the results will be significantly improved and latest configs and findings will be shared. Please wat...

FFurkan Gözükara SECourses https://youtu.be/nySGu12Y05k

G

GomezBroOP•9/3/24, 9:07 AM

I thought u did without captions on these newer lora trainings

GGomezBro I thought u did without captions on these newer lora trainings

F

Furkan Gözükara SECourses•9/3/24, 9:23 AM

yes i didnt use

F

Furkan Gözükara SECourses•9/3/24, 9:23 AM

it reduces likeliness of a person

F

Furkan Gözükara SECourses•9/3/24, 9:24 AM

currently for style i am testing

M

Maxivi•9/3/24, 9:24 AM

I have figured out regularzation images

M

Maxivi•9/3/24, 9:24 AM

still testing, but quite close

F

Furkan Gözükara SECourses•9/3/24, 9:39 AM

early checkpoints arrived

F

Furkan Gözükara SECourses•9/3/24, 9:39 AM

https://huggingface.co/MonsterMMORPG/3D-Cartoon-Style-FLUX/tree/main

MonsterMMORPG/3D-Cartoon-Style-FLUX at main

F

Furkan Gözükara SECourses•9/3/24, 10:06 AM

looks like LR is too low. 125 epochs and not learnt

F

Furkan Gözükara SECourses•9/3/24, 10:11 AM

maybe i am wrong looks like captioned one works

F

Furkan Gözükara SECourses•9/3/24, 10:11 AM

with lora

F

Furkan Gözükara SECourses•9/3/24, 10:11 AM

1008-a20ohwx203d20rendering20of20a20treasure20chest-flux1-dev-1762083518.png

F

Furkan Gözükara SECourses•9/3/24, 10:11 AM

left org - right lora

F

Furkan Gözükara SECourses•9/3/24, 10:19 AM

H

Hastur•9/3/24, 10:33 AM

Very nice results!

HHastur Very nice results!

F

Furkan Gözükara SECourses•9/3/24, 10:37 AM

yep

V

virginia•9/3/24, 10:39 AM

@Dr. Furkan Gözükara morning, could you double check this, =10:38:16.887 [Info] Self-Start ComfyUI-0 unexpectedly exited (if something failed, change setting LogLevel to Debug to see why!)
10:38:16.887 [Info] Self-Start ComfyUI-0 had errors before shutdown:
[ComfyUI-0/STDERR] Traceback (most recent call last):
[ComfyUI-0/STDERR] File "/kaggle/working/SwarmUI/dlbackend/ComfyUI/main.py", line 87, in <module>
[ComfyUI-0/STDERR] import comfy.utils
[ComfyUI-0/STDERR] File "/kaggle/working/SwarmUI/dlbackend/ComfyUI/comfy/utils.py", line 20, in <module>
[ComfyUI-0/STDERR] import torch
[ComfyUI-0/STDERR] File "/kaggle/working/SwarmUI/dlbackend/ComfyUI/venv/lib/python3.10/site-packages/torch/init.py", line 290, in <module>
[ComfyUI-0/STDERR] from torch._C import * # noqa: F403
[ComfyUI-0/STDERR] ImportError: /kaggle/working/SwarmUI/dlbackend/ComfyUI/venv/lib/python3.10/site-packages/torch/lib/../../nvidia/cusparse/lib/libcusparse.so.12: undefined symbol: __nvJitLinkComplete_12_4, version libnvJitLink.so.12, I have experienced a lot these days

Vvirginia @Dr. Furkan Gözükara morning, could you double check this, =10:38:16.887 [Info]...

F

Furkan Gözükara SECourses•9/3/24, 10:40 AM

kaggle broken due to comfyui update

F

Furkan Gözükara SECourses•9/3/24, 10:40 AM

i reported to both comfyui developer

F

Furkan Gözükara SECourses•9/3/24, 10:40 AM

and pytorch

F

Furkan Gözükara SECourses•9/3/24, 10:40 AM

https://github.com/pytorch/pytorch/issues/134929

GitHub

On Kaggle : libcusparse.so.12: undefined symbol: __nvJitLinkComplet...

Describe the bug I have tried everything but no luck Waiting your inputs to try more I tried torch 2.4.0, 2.5 - dev, cu 118, cu121 and cu124 - all same error This below code - I got the same erro...

F

Furkan Gözükara SECourses•9/3/24, 10:40 AM

no solution yet

T

Tacite•9/3/24, 11:00 AM

@Dr. Furkan Gözükara I just downloaded your JoyCaption V12 zip file. Inside is the V11 and also got the NumPy error

TTacite @Dr. Furkan Gözükara I just downloaded your JoyCaption V12 zip file. Inside is t...

F

Furkan Gözükara SECourses•9/3/24, 11:03 AM

pythonm mismatch very likely

F

Furkan Gözükara SECourses•9/3/24, 11:03 AM

what is ypur default python

T

Tacite•9/3/24, 11:04 AM

it works fine it's just a warning

V

virginia•9/3/24, 11:04 AM

okay thanks a lot

T

Tacite•9/3/24, 11:04 AM

I have 3.10.14

T

Tacite•9/3/24, 11:05 AM

ah no, it says finished but theres no file

TTacite ah no, it says finished but theres no file

F

Furkan Gözükara SECourses•9/3/24, 11:06 AM

check outputs folder

TTacite I have 3.10.14

F

Furkan Gözükara SECourses•9/3/24, 11:06 AM

it should work

T

Tacite•9/3/24, 11:07 AM

Process Process-2:
Traceback (most recent call last):
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 183, in convert_to_tensors
    tensor = as_tensor(value)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 142, in as_tensor
    return torch.tensor(value)
RuntimeError: Could not infer dtype of numpy.float32

Process Process-2:
Traceback (most recent call last):
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 183, in convert_to_tensors
    tensor = as_tensor(value)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 142, in as_tensor
    return torch.tensor(value)
RuntimeError: Could not infer dtype of numpy.float32

Process Process-2:
Traceback (most recent call last):
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 183, in convert_to_tensors
    tensor = as_tensor(value)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 142, in as_tensor
    return torch.tensor(value)
RuntimeError: Could not infer dtype of numpy.float32

Process Process-2:
Traceback (most recent call last):
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 183, in convert_to_tensors
    tensor = as_tensor(value)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 142, in as_tensor
    return torch.tensor(value)
RuntimeError: Could not infer dtype of numpy.float32

T

Tacite•9/3/24, 11:07 AM

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Program Files\Python310\lib\multiprocessing\process.py", line 314, in _bootstrap
    self.run()
  File "C:\Program Files\Python310\lib\multiprocessing\process.py", line 108, in run
    self._target(*self._args, **self._kwargs)
  File "E:\Joy_Caption_v12\app.py", line 335, in process_gpu_batch
    images = clip_processor(images=batch_images, return_tensors='pt', padding=True).pixel_values.to(device)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\models\siglip\processing_siglip.py", line 113, in __call__
    image_features = self.image_processor(images, return_tensors=return_tensors)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\image_processing_utils.py", line 41, in __call__
    return self.preprocess(images, **kwargs)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\models\siglip\image_processing_siglip.py", line 259, in preprocess
    return BatchFeature(data=data, tensor_type=return_tensors)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 79, in __init__
    self.convert_to_tensors(tensor_type=tensor_type)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 189, in convert_to_tensors
    raise ValueError(
ValueError: Unable to create tensor, you should probably activate padding with 'padding=True' to have batched tensors with the same length.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Program Files\Python310\lib\multiprocessing\process.py", line 314, in _bootstrap
    self.run()
  File "C:\Program Files\Python310\lib\multiprocessing\process.py", line 108, in run
    self._target(*self._args, **self._kwargs)
  File "E:\Joy_Caption_v12\app.py", line 335, in process_gpu_batch
    images = clip_processor(images=batch_images, return_tensors='pt', padding=True).pixel_values.to(device)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\models\siglip\processing_siglip.py", line 113, in __call__
    image_features = self.image_processor(images, return_tensors=return_tensors)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\image_processing_utils.py", line 41, in __call__
    return self.preprocess(images, **kwargs)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\models\siglip\image_processing_siglip.py", line 259, in preprocess
    return BatchFeature(data=data, tensor_type=return_tensors)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 79, in __init__
    self.convert_to_tensors(tensor_type=tensor_type)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 189, in convert_to_tensors
    raise ValueError(
ValueError: Unable to create tensor, you should probably activate padding with 'padding=True' to have batched tensors with the same length.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Program Files\Python310\lib\multiprocessing\process.py", line 314, in _bootstrap
    self.run()
  File "C:\Program Files\Python310\lib\multiprocessing\process.py", line 108, in run
    self._target(*self._args, **self._kwargs)
  File "E:\Joy_Caption_v12\app.py", line 335, in process_gpu_batch
    images = clip_processor(images=batch_images, return_tensors='pt', padding=True).pixel_values.to(device)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\models\siglip\processing_siglip.py", line 113, in __call__
    image_features = self.image_processor(images, return_tensors=return_tensors)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\image_processing_utils.py", line 41, in __call__
    return self.preprocess(images, **kwargs)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\models\siglip\image_processing_siglip.py", line 259, in preprocess
    return BatchFeature(data=data, tensor_type=return_tensors)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 79, in __init__
    self.convert_to_tensors(tensor_type=tensor_type)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 189, in convert_to_tensors
    raise ValueError(
ValueError: Unable to create tensor, you should probably activate padding with 'padding=True' to have batched tensors with the same length.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Program Files\Python310\lib\multiprocessing\process.py", line 314, in _bootstrap
    self.run()
  File "C:\Program Files\Python310\lib\multiprocessing\process.py", line 108, in run
    self._target(*self._args, **self._kwargs)
  File "E:\Joy_Caption_v12\app.py", line 335, in process_gpu_batch
    images = clip_processor(images=batch_images, return_tensors='pt', padding=True).pixel_values.to(device)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\models\siglip\processing_siglip.py", line 113, in __call__
    image_features = self.image_processor(images, return_tensors=return_tensors)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\image_processing_utils.py", line 41, in __call__
    return self.preprocess(images, **kwargs)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\models\siglip\image_processing_siglip.py", line 259, in preprocess
    return BatchFeature(data=data, tensor_type=return_tensors)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 79, in __init__
    self.convert_to_tensors(tensor_type=tensor_type)
  File "E:\Joy_Caption_v12\venv\lib\site-packages\transformers\feature_extraction_utils.py", line 189, in convert_to_tensors
    raise ValueError(
ValueError: Unable to create tensor, you should probably activate padding with 'padding=True' to have batched tensors with the same length.

T

Tacite•9/3/24, 11:08 AM

i will try to downgrade numpy

TTacite ```Process Process-2: Traceback (most recent call last): File "E:\Joy_Caption_...

F

Furkan Gözükara SECourses•9/3/24, 11:08 AM

let me reinstlal

F

Furkan Gözükara SECourses•9/3/24, 11:08 AM

this shouldnt happen

T

Tacite•9/3/24, 11:09 AM