Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature request] Deploy without Internet #1368

Open
JueLuo99 opened this issue Aug 12, 2024 · 1 comment
Open

[Feature request] Deploy without Internet #1368

JueLuo99 opened this issue Aug 12, 2024 · 1 comment
Labels
enhancement New feature or request

Comments

@JueLuo99
Copy link

The current version requires an Internet connection to download the models when it is first used after deployment.

Will the future add ways to deploy without the Internet? This will make LibrePhotos a completely offline usable project.

Can I download and place these models in the specified location manually?

I didn't find where to download and place them in the documentation.

Thanks for any help.

@JueLuo99 JueLuo99 added the enhancement New feature or request label Aug 12, 2024
@derneuere
Copy link
Member

derneuere commented Aug 14, 2024

To enable LibrePhotos to work completely offline by manually downloading and placing the required machine learning models, you can follow the steps outlined below. This guide assumes that you have access to the Internet initially to download the models and then configure LibrePhotos to run in an offline environment.

Step 1: Download the Models Manually

Manually download the necessary models from their respective URLs. Below is a list of models used by LibrePhotos, along with their download links:

  1. im2txt (Captioning)
    • URL: https://github.com/LibrePhotos/librephotos-docker/releases/download/0.1/im2txt.tar.gz
  2. clip-embeddings (CLIP)
    • URL: https://github.com/LibrePhotos/librephotos-docker/releases/download/0.1/clip-embeddings.tar.gz
  3. places365 (Categories)
    • URL: https://github.com/LibrePhotos/librephotos-docker/releases/download/0.1/places365.tar.gz
  4. resnet18 (Categories)
    • URL: https://download.pytorch.org/models/resnet18-5c106cde.pth
  5. im2txt_onnx (Captioning)
    • URL: https://github.com/LibrePhotos/librephotos-docker/releases/download/0.1/im2txt_onnx.tar.gz
  6. blip_base_capfilt_large (Captioning) (optional)
    • URL: https://huggingface.co/derneuere/librephotos_models/resolve/main/blip_large.tar.gz?download=true
  7. mistral-7b-v0.1.Q5_K_M (LLM) (optional)
    • URL: https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/resolve/main/mistral-7b-v0.1.Q5_K_M.gguf?download=true
  8. mistral-7b-instruct-v0.2.Q5_K_M (LLM) (optional)
    • URL: https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/resolve/main/mistral-7b-instruct-v0.2.Q5_K_M.gguf?download=true

Step 2: Place the Models in the Correct Location

Once the models are downloaded, place them in the following directory:

<LibrePhotos Media Root Directory>/data_models/

For example, if your MEDIA_ROOT is set to /var/lib/librephotos, then the models should be placed in /var/lib/librephotos/data_models/.

  • im2txt.tar.gz -> Unpack into <MEDIA_ROOT>/data_models/im2txt/
  • clip-embeddings.tar.gz -> Unpack into <MEDIA_ROOT>/data_models/clip-embeddings/
  • places365.tar.gz -> Unpack into <MEDIA_ROOT>/data_models/places365/
  • resnet18-5c106cde.pth -> Place directly as <MEDIA_ROOT>/data_models/resnet18-5c106cde.pth
  • im2txt_onnx.tar.gz -> Unpack into <MEDIA_ROOT>/data_models/im2txt_onnx/
  • blip_large.tar.gz -> Unpack into <MEDIA_ROOT>/data_models/blip/
  • mistral-7b-v0.1.Q5_K_M.gguf -> Place directly as <MEDIA_ROOT>/data_models/mistral-7b-v0.1.Q5_K_M.gguf
  • mistral-7b-instruct-v0.2.Q5_K_M.gguf -> Place directly as <MEDIA_ROOT>/data_models/mistral-7b-instruct-v0.2.Q5_K_M.gguf

Step 3: Verify Model Placement

Ensure that all models are correctly placed and unpacked in their respective directories. The structure should look something like this:

data_models/
    ├── im2txt/
    ├── clip-embeddings/
    ├── places365/
    ├── resnet18-5c106cde.pth
    ├── im2txt_onnx/
    ├── blip/
    ├── mistral-7b-v0.1.Q5_K_M.gguf
    └── mistral-7b-instruct-v0.2.Q5_K_M.gguf

Step 4: Run LibrePhotos

You can now run LibrePhotos without an active Internet connection. The application will use the models you manually downloaded and placed in the data_models directory.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

2 participants