..  _Quick start:

Quick start
********************************************************************************


Installation
------------------------------------------------

MultiNMRFit requires Python 3.8 or higher. If you do not have a Python environment
configured on your computer, we recommend that you follow the instructions
from `Anaconda <https://www.anaconda.com/download/>`_.

Then, open a terminal (e.g. run *Anaconda Prompt* if you have installed Anaconda) and type:

.. code-block:: bash

  pip install multinmrfit

You are now ready to start MultiNMRFit.

If this method does not work, you should ask your local system administrator or
the IT department "how to install a Python 3 package from PyPi" on your computer.

Alternatives & update
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

If you know that you do not have permission to install software systemwide,
you can install MultiNMRFit into your user directory using the :samp:`--user` flag:

.. code-block:: bash

  pip install --user multinmrfit


If you already have a previous version of MultiNMRFit installed, you can upgrade it to the latest version with:

.. code-block:: bash

  pip install --upgrade multinmrfit


Alternatively, you can also download all sources in a tarball from `GitHub <https://github.com/NMRTeamTBI/MultiNMRFit>`_,
but it will be more difficult to update MultiNMRFit later on.


Usage
------------------------------------------------

Graphical User Interface
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

To start the Graphical User Interface, type in a terminal (Windows: *Anaconda Prompt*):

.. code-block:: bash

  nmrfit

The MultiNMRFit window will open. If the window fails to open, have a look at our
:ref:`dedicated troubleshooting procedure <failed_gui>` to solve the problem.

.. image:: _static/interface.jpg
  :scale: 60%

The main processing steps can be performed via the menu on the left side bar:

* **Inputs & Outputs**: information required to load the data to process and export results (type of data, input and output directories, etc)
* **Process spectra**: process one or several signal(s) of specific spectra
* **Process from reference**: process a serie of spectra as done on a given spectrum (used as reference)
* **Results visualisation**: view and export processing results

Details on MultiNMRFit usage can be found in the :ref:`tutorial` section.

.. note:: The process is continuously and automatically saved as a pickle file in the output folder. To reopen the current processing state, just reopen this file by clicking on "Load a processing file - Browse files" on the side bar at the left.

.. warning:: MultiNMRFit silently overwrites (results and processing) files if they already exist. So take care to copy your results elsewhere or to change the output path and/or filename if you want to protect them from overwriting.

Library
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

MultiNMRFit is also available as a library (a Python module) that you can import directly in your Python
scripts:

.. code-block:: python

  import multinmrfit

.. seealso::  Have a look at our :ref:`API <Library documentation>` if you are interested in this feature.