Comments (6)
This sound like something is not quite right with the folders used... This can especially happen for dreambooth. Was this while trying to do dreambooth?
from kohya_ss.
Dreambooth Lora training.
from kohya_ss.
Try using the tool to prepare the folder structure for dreambooth. This will ensure all is proper for the training as far as folder names and all is concerned.
from kohya_ss.
I did. Many times, but still getting same error. I didn't have that problem few days ago. I was able to start training.
accelerate launch --num_cpu_threads_per_process=16 "train_network.py" --pretrained_model_name_or_path="C:/Programs/stable-diffusion-webui/models/Stable-diffusion/sd-v1-4.ckpt" --train_data_dir="C:/train_data/100_wny style" --resolution=512,512 --output_dir="C:/train_data" --use_8bit_adam --xformers --logging_dir="" --network_module=networks.lora --text_encoder_lr=5e-5 --unet_lr=1e-3 --network_dim=8 --output_name="last" --learning_rate="1e-5" --lr_scheduler="cosine" --train_batch_size="1" --save_every_n_epochs="1" --mixed_precision="fp16" --save_precision="fp16" --seed="1234" --cache_latents --max_data_loader_n_workers="1" --gradient_checkpointing --xformers --use_8bit_adam
prepare tokenizer
Use DreamBooth method.
prepare train images.
0 train images with repeating.
loading image sizes.
0it [00:00, ?it/s]
prepare dataset
No data found. Please verify arguments / 画像がありません。引数指定を確認してください
from kohya_ss.
Found the solution.
Interesting. It works only, when I point to the folder where the folder with images is located.
So instead of:
/img/100_wny style
I choose:
/img
There is no images inside of folder /img. They are located only in /100_wny style folder.
So I don't understand what happened and before I didn't have to do this trick. Weird.
Folder 100_wny style: 700 steps
max_train_steps = 700
stop_text_encoder_training = 0
lr_warmup_steps = 70
accelerate launch --num_cpu_threads_per_process=16 "train_network.py" --pretrained_model_name_or_path="C:/Programs/stable-diffusion-webui/models/Stable-diffusion/sd-v1-4.ckpt" --train_data_dir="C:/Users/pdept/Desktop/AI pics/training/train data/img" --resolution=512,512 --output_dir="C:/Users/pdept/Desktop/AI pics/training/train data/model" --use_8bit_adam --xformers --logging_dir="" --network_module=networks.lora --text_encoder_lr=5e-5 --unet_lr=1e-3 --network_dim=8 --output_name="wny" --learning_rate="1e-5" --lr_scheduler="cosine" --lr_warmup_steps="70" --train_batch_size="1" --max_train_steps="700" --save_every_n_epochs="1" --mixed_precision="fp16" --save_precision="fp16" --seed="1234" --cache_latents --max_data_loader_n_workers="1" --xformers --use_8bit_adam
prepare tokenizer
Use DreamBooth method.
prepare train images.
found directory 100_wny style contains 7 image files
700 train images with repeating.
loading image sizes.
100%|███████████████████████████████████████████████████████████████████████████████████| 7/7 [00:00<00:00, 202.55it/s]
prepare dataset
prepare accelerator
Using accelerator 0.15.0 or above.
load StableDiffusion checkpoint
loading u-net:
loading vae:
from kohya_ss.
Same issue and solution for me.
from kohya_ss.
Related Issues (20)
- After Installation Memory Leak HOT 1
- Parse safetensors files into a JSON configuration file
- is it possible to stop text encoder training after a certain amount of steps or epochs for sdxl dreambooth.
- Please adopt conda instead of env HOT 2
- Running tensorboard prompts Error: logging folder does not exist or does not contain logs. HOT 7
- returned non-zero exit status 3221225477. HOT 3
- BLIP 2 Caption Issue
- Problem while installing: error: metadata-generation-failed HOT 6
- Docker Compose not functional HOT 1
- Can't install - Gradio won't install HOT 3
- Subject: Issue with Loading State Dictionary for UNet2DConditionModel HOT 2
- Having problems training with Error: ValueError: torch.cuda.is_available() HOT 6
- How do i fix this error?
- "Stop TE (% of total steps)" being passed as-is (dreambooth SD1)
- Cant open bat file HOT 11
- Training issues HOT 1
- Model train causes TypeError issue HOT 10
- Flux.1 LoRA training HOT 246
- kohya_ss无法训练 HOT 9
- Please Help with sudden error while training lora model! HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from kohya_ss.