Skip to content

Commit

Permalink
Update installation documentation
Browse files Browse the repository at this point in the history
  • Loading branch information
tayden committed Jan 31, 2025
1 parent 5171c59 commit 9a0301a
Showing 1 changed file with 80 additions and 23 deletions.
103 changes: 80 additions & 23 deletions docs/installation.md
Original file line number Diff line number Diff line change
@@ -1,53 +1,110 @@
# Installation and Updating

The library is currently available for Python versions 3.9 through 3.11. Support for
future versions
will be added when possible.
## Requirements

New versions of the tool are occasionally released to improve segmentation performance,
speed, and the user interface of the tool. Changes are published to the PyPI and
Anaconda repositories using [semantic versioning](https://semver.org/). You may want to
occasionally run the update commands to ensure that you're using the most up-to-date
version of `kelp-o-matic`.
Kelp-o-Matic is currently available for Python versions 3.10 through 3.13.

## Commands
## Quick Install

=== "PIP"

```bash
pip install kelp-o-matic
```

=== "Conda"

```bash
conda install -c conda-forge kelp-o-matic
```

!!! help "Need more help?"

If you are unfamiliar with the command line or installing Python packages, you may
find our [Beginner Guide](beginner_guide/index.md) helpful.

### Pre-requisites
## Recommended Installation (Virtual Environment)

Install `pytorch` and `torchvision` for your operating system using the
[official installation instructions here](https://pytorch.org/).
Make sure you select "CUDA" as the compute platform if you have an NVIDIA GPU you'd like
to use to improve performance.
It is recommended to install Kelp-o-Matic in a virtual environment to avoid conflicts
with other Python packages present on your system. `uv` is our recommended environment manager
(a faster alternative to `venv`/`virtualenv`), but you can use any Python environment manager you prefer.

=== "PIP"
**UV Virtual Environment Setup**

### Install
1. Install `uv` using their [installation instructions](https://docs.astral.sh/uv/getting-started/installation/).

2. Create a new virtual environment (you can replace `kom-env` with a different name if you prefer):
```bash
pip install kelp-o-matic
uv venv kom-env
```

### Update
3. Activate the virtual environment

=== "Windows"
```powershell
kom-env\Scripts\activate
```

=== "MacOS/Linux"
```bash
source kom-env/bin/activate
```

4. Install Kelp-o-Matic and its dependencies

```bash
pip install --upgrade kelp-o-matic
uv pip install kelp-o-matic
```

=== "Conda"
5. If you want to deactivate the virtual environment, run

```bash
deactivate
```

It will deactivate automatically when you close the terminal. To reactivate, just run `source kom-env/bin/activate`
(you don't have to reinstall Kelp-o-Matic though).
!!! important
The `kom` command will be available only when the virtual environment is activated.
## Verify Installation
To verify that Kelp-o-Matic was installed correctly and to check the installed version, run:
```bash
kom --version
```
If you have an Nvidia GPU, you can check if it is detected with:
```bash
kom --gpu-test
```
### Install
## Updating
We follow [semantic versioning](https://semver.org/). Regular updates are recommended for:
- Performance improvements
- New features
- Security patches
To update Kelp-o-Matic, you can use the following commands:
=== "PIP"
```bash
conda install -c conda-forge kelp-o-matic
pip install --upgrade kelp-o-matic
```
### Update
=== "Conda"
```bash
conda update -c conda-forge kelp-o-matic
```
## Next Steps
For available commands, see the [Command Line Interface](cli.md) documentation.

0 comments on commit 9a0301a

Please sign in to comment.