mudler
Image Generation: ModuleNotFoundError: No module named 'PIL'
LocalAI version:
Container Image: quay.io/go-skynet/local-ai:v2.14.0-cublas-cuda12-ffmpeg-core
Environment, CPU architecture, OS, and Version:
Running on K8s
Describe the bug
Downloaded Dreamshaper via WebUI. Then wanted to use WebUI for Text-to-image, which gives me this error in logs.
To Reproduce
- Download Dreamshaper
- Use Text-2-Image
Expected behavior
It just works
Logs
│ localai-local-ai 6:17AM DBG GRPC(DreamShaper_8_pruned.safetensors-127.0.0.1:37247): stderr /build/backend/python/diffusers/run.sh: line 13: activate: No such file or directory │
│ localai-local-ai 6:17AM DBG GRPC(DreamShaper_8_pruned.safetensors-127.0.0.1:37247): stderr Traceback (most recent call last): │
│ localai-local-ai 6:17AM DBG GRPC(DreamShaper_8_pruned.safetensors-127.0.0.1:37247): stderr File "/build/backend/python/diffusers/backend_diffusers.py", line 12, in <module> │
│ localai-local-ai 6:17AM DBG GRPC(DreamShaper_8_pruned.safetensors-127.0.0.1:37247): stderr from PIL import Image │
│ localai-local-ai 6:17AM DBG GRPC(DreamShaper_8_pruned.safetensors-127.0.0.1:37247): stderr ModuleNotFoundError: No module named 'PIL'
Additional context
Hello Nold360, it seems there's a missing module in your LocalAI environment. To fix the issue, you need to install the 'PIL' module. You can do so by running the command 'pip install PIL'. If you encounter any issues while installing, please let us know and we'll assist you further. Also, this is an experiment by @mudler and I am a bot assisting with the audit of Github tickets. If you have any other questions or concerns, feel free to ask.
To fix the issue, you need to install the 'PIL' module. You can do so by running the command 'pip install PIL'.
Easier said then done in a container :P
@Nold360 for the python backends you need to pick up images without the core tag - the images with core are without any python dependencies, see also: https://localai.io/basics/container/#standard-container-images
