Replies: 1 comment
-
Hi, this probably means that the path you're using is not correct, also this will only work with powershell, you can test if the env variable is set correctly with this:
it should output: Alternatively you can just skip the use of env variables an use the path directly in the script like this:
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Ok my proplem is this i try to train with train_dreambooth_lora_sdxl.py. I am running it like this.
accelerate launch train_dreambooth_lora_sdxl.py --pretrained_model_name_or_path=$MODEL_NAME --instance_data_dir=$INSTANCE_DIR --pretrained_vae_model_name_or_path=$VAE_PATH --output_dir=$OUTPUT_DIR --mixed_precision="fp16" --instance_prompt="ther is text to" --resolution=1024 --train_batch_size=1 --gradient_accumulation_steps=4 --learning_rate=1e-4 --lr_scheduler="constant" --lr_warmup_steps=0 --max_train_steps=500 --validation_prompt="ther is text" --validation_epochs=25 --seed="0" --enable_xformers_memory_efficient_attention --gradient_checkpointing --use_8bit_adam
The env's are set like this
$env:INSTANCE_DIR="path to the folder with the images"
$env:VAE_PATH="path to the vae with .saftensor extension"
$env:OUTPUT_DIR= "path to the out put dir"
$env:MODEL_NAME=" path to the base model"
Normal that should work but if i run it i get my system can not found the path
env normal works for me on windows same as set in windows because windows has no export command flag
Beta Was this translation helpful? Give feedback.
All reactions