Metadata-Version: 2.1
Name: mpoints
Version: 1.0.0
Summary: Calibration and simulation of state-dependent Hawkes processes.
Author-email: Maxime Morariu-Patrichi <maximemorariu@me.com>, Claudio Bellani <claudio.bellani01@gmail.com>
Description-Content-Type: text/markdown
License-File: LICENSE

# The mpoints package

The `mpoints` package is a machine learning tool that implements the class of state-dependent Hawkes processes.
Its key features include both simulation and estimation (statistical inference).
It also contains a module with specialised plotting services.

State-dependent Hawkes processes belong to the class of hybrid marked point processes,
a class that models the arrival in time of random events and their interaction with the state of a system.

We strongly recommend to first read the [tutorial](https://mpoints.readthedocs.io/en/latest/tutorial.html).
It contains an introduction to this statistical model and illustrates the main services offered by the `mpoints` package.

For additional mathematical details, please consult the [documentation](https://mpoints.readthedocs.io) and the 
[references](https://mpoints.readthedocs.io/en/latest/references.html).

## Installation

The package can be easily installed via `pip`, a popular package management system for Python. In the terminal, simply enter the following command.

```
pip install mpoints
```
If you are using virtual environments (with conda), make sure that you install the package in the intended environment.

Note: when installing `mpoints` on Windows, you may be prompted to install Visual C++ Build Tools if this is not already installed.


## Build from source
#### Build the wheel using `setuptools`. 
```
pip install -q build
python -m build
```


#### Install in development mode
If you are contributing to `mpionts`, you might want to install it in development mode.
See [development-mode](https://setuptools.pypa.io/en/latest/userguide/quickstart.html#development-mode)

```
pip install --upgrade --force-reinstall --editable <path-to-mpoints-root>
```
