Metadata-Version: 2.4
Name: asreml
Version: 1.0.0a320
Summary: A statistical package that fits linear mixed models using Residual Maximum Likelihood (REML)
Author-email: The VSNi Team <asreml@vsni.co.uk>
Maintainer-email: The VSNi Team <asreml@vsni.co.uk>
Project-URL: homepage, https://vsni.co.uk/
Requires-Python: >=3.8
Description-Content-Type: text/x-rst
Requires-Dist: pandas
Requires-Dist: numpy
Requires-Dist: dotenv
Requires-Dist: requests
Provides-Extra: plots
Requires-Dist: plotly; extra == "plots"
Requires-Dist: scipy; extra == "plots"
Provides-Extra: static-images
Requires-Dist: asreml[plots]; extra == "static-images"
Requires-Dist: kaleido>=1.0.0; extra == "static-images"
Provides-Extra: full
Requires-Dist: asreml[plots,static_images]; extra == "full"

# asreml

asreml is a statistical package that fits linear mixed models using Residual Maximum Likelihood (REML).

## Features
- a flexible syntax for specifying variance models for the random effects, and
- the REML routines use the Average Information (AI) algorithm, and sparse matrix methods for fitting the mixed model.
These enable asreml to fit many types of models and efficiently analyse large and complex data sets.


## Installation
### Linux
Note: the following instructions assume the use of bash.
#### Install the uv Python package and project manager.
[Docs](https://docs.astral.sh/uv): 
```
curl -LsSf https://astral.sh/uv/install.sh | sh
```
#### Install multiple python versions
```
uv python install 3.10 3.11 3.12
```
#### Create a virtual environment
```
uv venv --python 3.12 my_venv 
source . my_venv/bin/activate
```

#### Install
- Basic installation
```
my_venv/bin/pip install <PACKAGE_FILE>
```
- With plotting support
```
my_venv/bin/pip install <PACKAGE_FILE>[plots]
```
- Full installation
```
my_venv/bin/pip install <PACKAGE_FILE>[full]
```
where ```<package file>``` is the name of the .tar.gz or wheel file holding the package.

#### Activate License
If license activation is required, run the following command:
```
my_venv/bin/python <PACKAGE_PATH>/get_license.py YYYY-YYYY-YYYY-YYYY
```
where ```<PACKAGE_PATH>``` can be found using ```uv pip show asreml``` and ```YYYY-YYYY-YYYY-YYYY``` is your license key.

If the license is not found, set the ```VSN_USER_LICPATH``` environment variable to the directory containing the license file:
```
export VSN_USER_LICPATH="<PATH_TO_LICENSE>"
```
where ```<PATH_TO_LICENSE>``` is the directory holding the license file.


#### Windows
Note: the following instructions assume the use of PowerShell.
#### Install the uv Python package and project manager.
[Docs](https://docs.astral.sh/uv):
```
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
```
#### Create a virtual environment (optional)
```
uv venv my_venv
. \my_venv\bin\Activate.psl
```

#### Install
- Basic installation
```
my_venv/bin/pip install <PACKAGE_FILE>
```
- With plotting support
```
my_venv/bin/pip install <PACKAGE_FILE>[plots]
```
- Full installation
```
my_venv/bin/pip install <PACKAGE_FILE>[full]
```
where ```<package file>``` is the name of the .tar.gz or wheel file holding the package.

#### Activate License
If license activation is required, run the following command:
```
my_venv/bin/python <PACKAGE_PATH>\get_license.py YYYY-YYYY-YYYY-YYYY
```
where ```<PACKAGE_PATH>``` can be found using ```uv pip show asreml``` and ```YYYY-YYYY-YYYY-YYYY``` is your license key.


If the license is not found, set the ```VSN_USER_LICPATH``` environment variable to the directory containing the license file:
```
$env:VSN_USER_LICPATH = "<PATH_TO_LICENSE>"
```
where ```<PATH_TO_LICENSE>``` is the directory holding the license file.

# Change Log
All notable changes to this project will be documented in this file.

## [latest] - 2026-04-28

### Added

- zero(..) model function added
- twostageweights option has been added, and the subsequent output made
available

### Changed

- Default plotting colours changed

### Fixed

- All available predict tables will now be output, rather than stopping at the
first table that failed
- fa(..) model function now working as documented
- rr(..) model function now working as documented
- fixed an issue where models with a large number of levels in the fixed
effects (> 800) were hanging
