Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/abailoni/cellpose-training-server
https://github.com/abailoni/cellpose-training-server
Last synced: 24 days ago
JSON representation
- Host: GitHub
- URL: https://github.com/abailoni/cellpose-training-server
- Owner: abailoni
- License: mit
- Created: 2022-04-18T17:56:14.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-07-31T10:53:28.000Z (6 months ago)
- Last Synced: 2024-11-05T11:55:25.360Z (2 months ago)
- Language: Python
- Size: 1.14 MB
- Stars: 1
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Cellpose Training Server
Tool to train custom cellpose models via the browser on a remote server with GPU support, given cellpose training images generated using the [Cellpose Annotation Tool](https://git.embl.de/grp-alexandrov/cellpose-training-gui) tool.
## Installation and usage on the server with GPU support
### How to build the docker image
- Clone this repository with `git clone [email protected]:grp-alexandrov/cellpose-training-server.git`
- Move to the repository folder with `cd cellpose-training-server`
- To build the docker image, you will need to provide credentials to access the GitLab repository. Follow instructions [here](https://grp-alexandrov.embl-community.io/napari-spacem-qc/user-guide/access/local-installation.html#setting-up-private-package-registry) if you do not already have a `pip.config` file configured with your credentials. After that, you can use the tokens from your `pip.conf` file for the following command:
```
docker build -t cellposetrainingserverapp:latest --build-arg CI_DEPLOY_USER=__your_token__ --build-arg CI_DEPLOY_PASSWORD="your_password" ./dockerfiles
```
### How to start the docker image
- If you want to temporarily start the docker image, you can run the following command:
```shell
docker run -p 8501:8501 --gpus all cellposetrainingserverapp:latest
```
- If you want to start the docker image and keep it running in the background (even when your server is restarted), you can run the following command:
```shell
docker run -p 8501:8501 --gpus all --restart unless-stopped -d cellposetrainingserverapp:latest
```
Afterwards, you can access the tool via the browser at `http://localhost:8501`