GSoC 2021: Support of the simulation-based inference with the model fitting toolbox

.gitignore

@@ -80,3 +80,19 @@ target/
 
 # Installation and auto-generated files
 brian2modelfitting.egg-info
+
+# sbi logs
+examples/sbi-logs/
+brian2modelfitting/tests/sbi-logs/
+
+# other sbi and torch stuff
+examples/*.pt
+examples/*.pth
+brian2modelfitting/tests/*.pt
+brian2modelfitting/tests/*.pth
+
+# numpy files
+examples/*.npy
+examples/*.npz
+brian2modelfitting/tests/*.npy
+brian2modelfitting/tests/*.npz

README.md

@@ -1,28 +1,38 @@
 brian2modelfitting
 ==================
 
+Model fitting toolbox for Brian 2 simulator.
 
-Model Fitting Toolbox for Brian 2 simulator, to allow the user to find the best fit of the parameters for recorded traces and spike trains.
+[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.4601961.svg)](https://doi.org/10.5281/zenodo.4601961)
 
-By default, we support a range of global derivative-free optimization methods, that include popular methods for model fitting, such as: Differential Evolution, Particle Swarm Optimization and Covariance Matrix Adaptation (provided by the Nevergrad) as well as Bayesian Optimization for black box functions (provided by Scikit-Optimize).
 
-Documentation for Brian2modelfitting can be found at http://brian2modelfitting.readthedocs.org
 
-Brian2modelfitting is released under the terms of the CeCILL 2.1 license.
+The package [`brian2modelfitting`](https://brian2modelfitting.readthedocs.io) allows the user to find the best fit of the unknown free parameters for recorded traces and spike trains. It also supports simulation-based inference, where instead of point-estimated parameter values, a full posterior distribution over the parameters is computed.
 
-Please report issues at the github issue tracker (https://github.com/brian-team/brian2modelfitting/issues) or at the brian discussion forum (https://brian.discourse.group)
+By default, the toolbox supports a range of global derivative-free optimization methods, that include popular methods for model fitting: differential evolution, particle swarm optimization and covariance matrix adaptation (provided by the  [`Nevergrad`](https://facebookresearch.github.io/nevergrad/), a gradient-free optimization platform) as well as Bayesian optimization for black box functions (provided by [`scikit-optimize`](https://scikit-optimize.github.io/stable/), a sequential model-based optimization library). On the other hand, simulation-based inference is the process of finding parameters of a simulator from observations by taking a Bayesian approach via sequential neural posterior estimation, likelihood estimation or ration estimation (provided by the [`sbi`](https://www.mackelab.org/sbi/)), where neural densitiy estimator, a deep neural network allowing probabilistic association between the data and underlying parameter space, is trained. After the network is trained, the approximated posterior distribution is available.
 
-[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.4601961.svg)](https://doi.org/10.5281/zenodo.4601961)
+Documentation
+-------------
+[![Documentation Status](https://readthedocs.org/projects/ansicolortags/badge/?version=latest)](https://brian2modelfitting.readthedocs.io)
+
+The full documentation is available at http://brian2modelfitting.readthedocs.org.
+
+Testing status
+--------------
+[![Build Status](https://github.com/brian-team/brian2modelfitting/workflows/Tests/badge.svg)](https://github.com/brian-team/brian2modelfitting/actions) 
+[![Coverage Status](https://coveralls.io/repos/github/brian-team/brian2modelfitting/badge.svg?branch=master)](https://coveralls.io/github/brian-team/brian2modelfitting?branch=master)
 
 Installation
 ------------
-Module can be installed with pypi:
+Install `brian2modelfitting` from the Python package index via pip:
 ```
 pip install brian2modelfitting
 ```
 
+License
+-------
+The model fitting toolbox is released under the terms of the CeCILL 2.1 license and is available [here](https://github.com/brian-team/brian2modelfitting/blob/master/LICENSE).
 
-Testing status
---------------
-[![Build Status](https://github.com/brian-team/brian2modelfitting/workflows/Tests/badge.svg)](https://github.com/brian-team/brian2modelfitting/actions) 
-[![Coverage Status](https://coveralls.io/repos/github/brian-team/brian2modelfitting/badge.svg?branch=master)](https://coveralls.io/github/brian-team/brian2modelfitting?branch=master)
+Use
+---
+Please report issues at the GitHub [issue tracker](https://github.com/brian-team/brian2modelfitting/issues) or at the [Brian 2 discussion forum](https://brian.discourse.group).

brian2modelfitting/__init__.py

@@ -6,6 +6,7 @@
 
 from .tests import run as run_test
 from .fitter import *
+from .inferencer import Inferencer
 from .optimizer import *
 from .metric import *
 from .simulator import *
@@ -16,4 +17,5 @@
            'Simulator', 'RuntimeSimulator', 'CPPStandaloneSimulator',
            'MSEMetric', 'Metric', 'GammaFactor', 'FeatureMetric',
            'SpikeMetric', 'TraceMetric', 'get_gamma_factor', 'firing_rate',
-           'Fitter', 'SpikeFitter', 'TraceFitter', 'OnlineTraceFitter']
+           'Fitter', 'SpikeFitter', 'TraceFitter', 'OnlineTraceFitter',
+           'Inferencer']