John6666

/

joy-caption-alpha-two-cli-mod

like 20

Oh my god, why are there always so many problems with linux-like systems. Now my git is wrong:

Download new git:
https://github.com/git-lfs/git-lfs/releases

Correct path:
set PATH=%DIR%\git-lfs;%DIR%\python;%DIR%\python\Scripts;%PATH%

(venv) C:\python2\JoyCaption\joy-caption-alpha-two-cli-mod>git lfs install
Updated Git hooks.
Git LFS initialized.

(venv) C:\python2\JoyCaption\joy-caption-alpha-two-cli-mod>git lfs update
Updated Git hooks.

(venv) C:\python2\JoyCaption\joy-caption-alpha-two-cli-mod>git lfs pull
Downloading LFS objects: 33% (1/3), 1.2 GB | 3.6 MB/s

(venv) C:\python2\JoyCaption\joy-caption-alpha-two-cli-mod>pip install -r requirements.txt
...

(venv) C:\python2\JoyCaption\joy-caption-alpha-two-cli-mod>pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
...

(venv) C:\python2\JoyCaption\joy-caption-alpha-two-cli-mod>python app.py D:\Stable_Diffusion\pic\13
...
VLM's custom text model isn't loaded 🤖
Loading image adapter 🖼️
Processing directory 📁: D:\Stable_Diffusion\pic\13
Output directory 📦: D:\Stable_Diffusion\pic\13
Batch size 🗄️: 4
Processing images: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████| 12/12 [04:12<00:00, 21.08s/image]

(venv) C:\python2\JoyCaption\joy-caption-alpha-two-cli-mod>python app.py D:\Stable_Diffusion\pic\13
Running on cuda
Loading in NF4
Loading CLIP 📎
Loading VLM's custom vision model 📎
Error loading models: invalid load key, 'v'.

same here. Do you have any solutions?

This error occurs when the version of git lfs is old, and there is also another problem where the operation of git lfs itself fails if the version of git is old.
So, you need to update git first, and then update git lfs to resolve the issue.
For Windows, download and install from the link below.
https://git-scm.com/downloads/win
https://git-lfs.com/