Seg2Link

https://github.com/wenchentao/seg2link
Admin
Seg2Link is a napari-based tool for semi-automatic cell segmentation and manual refinement in 3D...
Tokens:10,377
Snippets:108
Trust Score:5.6
License:MIT
Update:3 weeks ago
Show doc for...
Tokens:
Raw
### Install seg2Link-unet2d with local extras

Source: https://github.com/wenchentao/seg2link/blob/master/docs/seg2link-unet2d.md

Install the seg2Link-unet2d package with the 'local' extra in the active unet2d environment. Requires prior installation of PyTorch.

```console
(unet2d) $ pip install seg2Link-unet2d[local]
```

--------------------------------

### Launch seg2link

Source: https://github.com/wenchentao/seg2link/blob/master/README.md

Starts the Seg2Link napari-based software after activating the environment.

```console
$ seg2link
```

--------------------------------

### Install Seg2Link with pip

Source: https://github.com/wenchentao/seg2link/blob/master/docs/quick_start.md

Installs the Seg2Link package using pip inside the activated 'seg2link-env' environment.

```console
(seg2link-env) $ pip install seg2link
```

--------------------------------

### Activate environment and launch Seg2Link

Source: https://github.com/wenchentao/seg2link/blob/master/docs/quick_start.md

Activates the 'seg2link-env' environment and launches the Seg2Link application. The command 'seg2link' starts the GUI.

```console
$ conda activate seg2link-env
(seg2link-env) $ seg2link
```

--------------------------------

### Activate seg2link environment

Source: https://github.com/wenchentao/seg2link/blob/master/README.md

Activates the seg2link-env conda environment before starting the software.

```console
$ conda activate seg2link-env
```

--------------------------------

### Change to notebook directory

Source: https://github.com/wenchentao/seg2link/blob/master/docs/seg2link-unet2d.md

Change to the directory containing the training notebook. Replace C:/xxx/xxx/ with the actual path.

```console
(unet2d) $ cd C:/xxx/xxx/
```

--------------------------------

### Local training directory structure

Source: https://github.com/wenchentao/seg2link/blob/master/docs/seg2link-unet2d.md

Directory structure for local training: the notebook and the unet_01 folder (containing raw_image, train_image, train_label) must be in the same parent directory.

```text
C:/xxx/xxx
    ├───Training_notebook_xxx.ipynb
    └───unet_01
        ├───raw_image
        │       ├───raw_image0000.tif
        │       ├───raw_image0001.tif
        │       ├───raw_image0002.tif
        │       └───...
        ├───train_image
        │       ├───train_image0000.tif
        │       ├───train_image0001.tif
        │       ├───train_image0002.tif
        │       └───...
        └───train_label
                ├───train_cells0000.tif
                ├───train_cells0001.tif
                ├───train_cells0002.tif
                └───...    
```

--------------------------------

### Launch Jupyter Notebook

Source: https://github.com/wenchentao/seg2link/blob/master/docs/seg2link-unet2d.md

Launch Jupyter Notebook from the terminal in the unet2d environment.

```console
(unet2d) $ jupyter notebook
```

--------------------------------

### Create and activate conda environment

Source: https://github.com/wenchentao/seg2link/blob/master/docs/quick_start.md

Creates a new conda environment named 'seg2link-env' with Python 3.8 and pip, then activates it. Run these commands in Anaconda PowerShell (Windows) or terminal (macOS/Linux).

```console
$ conda create -n seg2link-env python=3.8 pip
$ conda activate seg2link-env
```

--------------------------------

### Create and activate conda environment unet2d

Source: https://github.com/wenchentao/seg2link/blob/master/docs/seg2link-unet2d.md

Commands to create a new conda environment named 'unet2d' and activate it. Run in Anaconda PowerShell Prompt (Windows) or terminal (macOS/Linux).

```console
$ conda create -n unet2d
$ conda activate unet2d
```

--------------------------------

### Google Colab directory structure

Source: https://github.com/wenchentao/seg2link/blob/master/docs/seg2link-unet2d.md

Directory structure for Google Colab: upload the unet_01 folder (containing raw_image, train_image, train_label) to the root of your Google Drive.

```text
root of your Google drive
    └───unet_01
        ├───raw_image
        │       ├───raw_image0000.tif
        │       ├───raw_image0001.tif
        │       ├───raw_image0002.tif
        │       └───...
        ├───train_image
        │       ├───train_image0000.tif
        │       ├───train_image0001.tif
        │       ├───train_image0002.tif
        │       └───...
        └───train_label
                ├───train_cells0000.tif
                ├───train_cells0001.tif
                ├───train_cells0002.tif
                └───...    
```

--------------------------------

### Update seg2link to latest version

Source: https://github.com/wenchentao/seg2link/blob/master/README.md

Updates seg2link to the latest version using pip.

```console
$ pip install --upgrade seg2link
```
Tokens:
Raw