Open-source Python Toolbox Research Articles

A Framework for the Optimization of Complex Cyber-Physical Systems via Directed Acyclic Graph.

Mathematical modeling and data-driven methodologies are frequently required to optimize industrial processes in the context of Cyber-Physical Systems (CPS). This paper introduces the PipeGraph software library, an open-source python toolbox for easing the creation of machine learning models by using Directed Acyclic Graph (DAG)-like implementations that can be used for CPS. scikit-learn’s Pipeline is a very useful tool to bind a sequence of transformers and a final estimator in a single unit capable of working itself as an estimator. It sequentially assembles several steps that can be cross-validated together while setting different parameters. Steps encapsulation secures the experiment from data leakage during the training phase. The scientific goal of PipeGraph is to extend the concept of Pipeline by using a graph structure that can handle scikit-learn’s objects in DAG layouts. It allows performing diverse operations, instead of only transformations, following the topological ordering of the steps in the graph; it provides access to all the data generated along the intermediate steps; and it is compatible with GridSearchCV function to tune the hyperparameters of the steps. It is also not limited to entries. Moreover, it has been proposed as part of the scikit-learn-contrib supported project, and is fully compatible with scikit-learn. Documentation and unitary tests are publicly available together with the source code. Two case studies are analyzed in which PipeGraph proves to be essential in improving CPS modeling and optimization: the first is about the optimization of a heat exchange management system, and the second deals with the detection of anomalies in manufacturing processes.

AbspectroscoPY, a Python toolbox for absorbance-based sensor data in water quality monitoring

Absorbance-based sensors produce large raw attenuation datasets. We developed AbspectroscoPY, an open-source Python toolbox to implement semi-automated processing of these data and explore the full potential of high-frequency measurements.

FLOMPY: An Open-Source Toolbox for Floodwater Mapping Using Sentinel-1 Intensity Time Series

A new automatic, free and open-source python toolbox for the mapping of floodwater is presented. The output of the toolbox is a binary mask of floodwater at a user-specified time point within geographical boundaries. It exploits the high spatial (10m) and temporal (6 days per orbit over Europe) resolution of Sentinel-1 GRD intensity time series and is based on four processing steps. In the first step, a selection of Sentinel-1 images related to pre-flood (baseline) state and flood state is performed. In the second step, the preprocessing of the selected images is performed in order to create a co-registered stack with all the pre-flood and flood images. In the third step, a statistical temporal analysis is performed and a t-score map that represents the changes due to a flood event is calculated. Finally, in the fourth step, a classification procedure based on the t-score map is performed to extract the final flood map. A thorough analysis based on several flood events is presented to demonstrate the main benefits, limitations and the potential of the proposed methodology. The validation was performed using Copernicus Emergency Management Service (EMS) products. In all case studies, overall accuracies were higher than 0.95 with Kappa scores higher than 0.76. We believe that the end-user community can benefit by exploiting the flood maps of the proposed methodological pipeline by using the provided open-source toolbox.

Evaluation of three approaches for simulating 3-D time-domain electromagnetic data

SUMMARY We implemented and compared the implicit Euler time-stepping approach, the inverse Fourier transform-based approach and the Rational Arnoldi method for simulating 3-D transient electromagnetic data. We utilize the finite-element method with unstructured tetrahedral meshes for the spatial discretization supporting irregular survey geometries and anisotropic material parameters. Both, switch-on and switch-off current waveforms, can be used in combination with direct current solutions of Poisson problems as initial conditions. Moreover, we address important topics such as the incorporation of source currents and opportunities to simulate impulse as well as step response magnetic field data with all approaches for supporting a great variety of applications. Three examples ranging from simple to complex real-world geometries and validations against external codes provide insight into the numerical accuracy, computational performance and unique characteristics of the three applied methods. We further present an application of logarithmic Fourier transforms to convert transient data into the frequency domain. We made all approaches available in the open-source Python toolbox custEM, which previously supported only frequency-domain electromagnetic data. The object-oriented software implementation is suited for further elaboration on distinct modelling topics and the presented examples can serve for benchmarking other codes.

Gradient-based and wavelet-based compressed sensing approaches for highly undersampled tomographic datasets

Electron tomography is widely employed for the 3D morphological characterization at the nanoscale. In recent years, there has been a growing interest in analytical electron tomography (AET) as it is capable of providing 3D information about the elemental composition, chemical bonding and optical/electronic properties of nanomaterials. AET requires advanced reconstruction algorithms as the datasets often consist of a very limited number of projections. Total variation (TV)-based compressed sensing approaches were shown to provide high-quality reconstructions from undersampled datasets, but staircasing artefacts can appear when the assumption about piecewise constancy does not hold. In this paper, we compare higher-order TV and wavelet-based approaches for AET applications and provide an open-source Python toolbox, Pyetomo, containing 2D and 3D implementations of both methods. A highly sampled STEM-HAADF dataset of an Er-doped porous Si sample and a heavily undersampled STEM-EELS dataset of a Ge-rich GeSbTe (GST) thin film annealed at 450°C are used to evaluate the performance of the different approaches. We show that polynomial annihilation with order 3 (HOTV3) and the Bior4.4 wavelet outperform the classical TV minimization and the related Haar wavelet.

Tensorpac: An open-source Python toolbox for tensor-based phase-amplitude coupling measurement in electrophysiological brain signals.

Despite being the focus of a thriving field of research, the biological mechanisms that underlie information integration in the brain are not yet fully understood. A theory that has gained a lot of traction in recent years suggests that multi-scale integration is regulated by a hierarchy of mutually interacting neural oscillations. In particular, there is accumulating evidence that phase-amplitude coupling (PAC), a specific form of cross-frequency interaction, plays a key role in numerous cognitive processes. Current research in the field is not only hampered by the absence of a gold standard for PAC analysis, but also by the computational costs of running exhaustive computations on large and high-dimensional electrophysiological brain signals. In addition, various signal properties and analyses parameters can lead to spurious PAC. Here, we present Tensorpac, an open-source Python toolbox dedicated to PAC analysis of neurophysiological data. The advantages of Tensorpac include (1) higher computational efficiency thanks to software design that combines tensor computations and parallel computing, (2) the implementation of all most widely used PAC methods in one package, (3) the statistical analysis of PAC measures, and (4) extended PAC visualization capabilities. Tensorpac is distributed under a BSD-3-Clause license and can be launched on any operating system (Linux, OSX and Windows). It can be installed directly via pip or downloaded from Github (https://github.com/EtienneCmb/tensorpac). By making Tensorpac available, we aim to enhance the reproducibility and quality of PAC research, and provide open tools that will accelerate future method development in neuroscience.

Efficient exact inference for dynamical systems with noisy measurements using sequential approximate Bayesian computation.

MotivationApproximate Bayesian computation (ABC) is an increasingly popular method for likelihood-free parameter inference in systems biology and other fields of research, as it allows analyzing complex stochastic models. However, the introduced approximation error is often not clear. It has been shown that ABC actually gives exact inference under the implicit assumption of a measurement noise model. Noise being common in biological systems, it is intriguing to exploit this insight. But this is difficult in practice, as ABC is in general highly computationally demanding. Thus, the question we want to answer here is how to efficiently account for measurement noise in ABC.ResultsWe illustrate exemplarily how ABC yields erroneous parameter estimates when neglecting measurement noise. Then, we discuss practical ways of correctly including the measurement noise in the analysis. We present an efficient adaptive sequential importance sampling-based algorithm applicable to various model types and noise models. We test and compare it on several models, including ordinary and stochastic differential equations, Markov jump processes and stochastically interacting agents, and noise models including normal, Laplace and Poisson noise. We conclude that the proposed algorithm could improve the accuracy of parameter estimates for a broad spectrum of applications.Availability and implementationThe developed algorithms are made publicly available as part of the open-source python toolbox pyABC (https://github.com/icb-dcm/pyabc).Supplementary information Supplementary data are available at Bioinformatics online.

Coupled magnetic resonance and electrical resistivity tomography: An open-source toolbox for surface nuclear-magnetic resonance

Nuclear-magnetic resonance (NMR) is a powerful tool for groundwater system imaging. Ongoing developments in surface NMR, for example, multichannel devices, allow for investigations of increasingly complex subsurface structures. However, with the growing complexity of field cases, the availability of appropriate software to accomplish the in-depth data analysis becomes limited. The open-source Python toolbox coupled magnetic resonance and electrical resistivity tomography (COMET) provides the community with a software for modeling and inversion of complex surface NMR data. COMET allows the NMR parameters’ water content and relaxation time to vary in one dimension or two dimensions and accounts for arbitrary electrical resistivity distributions. It offers a wide range of classes and functions to use the software via scripts without in-depth programming knowledge. We validated COMET to existing software for a simple 1D example. We discovered the potential of COMET by a complex 2D case, showing 2D inversions using different approximations for the resistivity, including a smooth distribution from electrical resistivity tomography (ERT). The use of ERT-based resistivity results in similar water content and relaxation time images compared with using the original synthetic block resistivity. We find that complex inversion may indicate incorrect resistivity by non-Gaussian data misfits, whereas amplitude inversion shows well-fitted data, but leading to erroneous NMR models.

Facilitating open-science with realistic fMRI simulation: validation and application.

With advances in methods for collecting and analyzing fMRI data, there is a concurrent need to understand how to reliably evaluate and optimally use these methods. Simulations of fMRI data can aid in both the evaluation of complex designs and the analysis of data. We present fmrisim, a new Python package for standardized, realistic simulation of fMRI data. This package is part of BrainIAK: a recently released open-source Python toolbox for advanced neuroimaging analyses. We describe how to use fmrisim to extract noise properties from real fMRI data and then create a synthetic dataset with matched noise properties and a user-specified signal. We validate the noise generated by fmrisim to show that it can approximate the noise properties of real data. We further show how fmrisim can help researchers find the optimal design in terms of power. The fmrisim package holds promise for improving the design of fMRI experiments, which may facilitate both the pre-registration of such experiments as well as the analysis of fMRI data.

A Comparison of a One-Dimensional Finite Element Method and the Transfer Matrix Method for the Computation of Wind Music Instrument Impedance

This work presents a computation tool for the calculation of wind instrument input impedance in the context of linear planar wave propagation with visco-thermal losses. The originality of the approach lies in the usage of a specific and simple 1D finite element method (FEM). The popular Transfer Matrix Method (TMM) is also recalled and a seamless formulation is proposed which unifies the cases cylinders vs. cones. Visco-thermal losses, which are natural dissipation in the system, are not exactly taken into account by this method when arbitrary shapes are considered. The introduction of an equivalent radius leads to an approximation that we quantify using the FEM method. The equation actually solved by the TMM in this case is exhibited. The accuracy of the two methods (FEM and TMM) and the associated computation times are assessed and compared. Although the TMM is more efficient in lossless cases and for lossy cylinders, the FEM is shown to be more efficient when targeting a specific precision in the realistic case of a lossy trumpet. Some additional features also exhibit the robustness and flexibility of the FEM over the TMM. All the results of this article are computed using the open-source python toolbox OpenWind.

Gumpy: a Python toolbox suitable for hybrid brain–computer interfaces

Objective. The objective of this work is to present gumpy, a new free and open source Python toolbox designed for hybrid brain–computer interface (BCI). Approach. Gumpy provides state-of-the-art algorithms and includes a rich selection of signal processing methods that have been employed by the BCI community over the last 20 years. In addition, a wide range of classification methods that span from classical machine learning algorithms to deep neural network models are provided. Gumpy can be used for both EEG and EMG biosignal analysis, visualization, real-time streaming and decoding. Results. The usage of the toolbox was demonstrated through two different offline example studies, namely movement prediction from EEG motor imagery, and the decoding of natural grasp movements with the applied finger forces from surface EMG (sEMG) signals. Additionally, gumpy was used for real-time control of a robot arm using steady-state visually evoked potentials (SSVEP) as well as for real-time prosthetic hand control using sEMG. Overall, obtained results with the gumpy toolbox are comparable or better than previously reported results on the same datasets. Significance. Gumpy is a free and open source software, which allows end-users to perform online hybrid BCIs and provides different techniques for processing and decoding of EEG and EMG signals. More importantly, the achieved results reveal that gumpy’s deep learning toolbox can match or outperform the state-of-the-art in terms of accuracy. This can therefore enable BCI researchers to develop more robust decoding algorithms using novel techniques and hence chart a route ahead for new BCI improvements.

Uncertainpy: A Python Toolbox for Uncertainty Quantification and Sensitivity Analysis in Computational Neuroscience

Computational models in neuroscience typically contain many parameters that are poorly constrained by experimental data. Uncertainty quantification and sensitivity analysis provide rigorous procedures to quantify how the model output depends on this parameter uncertainty. Unfortunately, the application of such methods is not yet standard within the field of neuroscience. Here we present Uncertainpy, an open-source Python toolbox, tailored to perform uncertainty quantification and sensitivity analysis of neuroscience models. Uncertainpy aims to make it quick and easy to get started with uncertainty analysis, without any need for detailed prior knowledge. The toolbox allows uncertainty quantification and sensitivity analysis to be performed on already existing models without needing to modify the model equations or model implementation. Uncertainpy bases its analysis on polynomial chaos expansions, which are more efficient than the more standard Monte-Carlo based approaches. Uncertainpy is tailored for neuroscience applications by its built-in capability for calculating characteristic features in the model output. The toolbox does not merely perform a point-to-point comparison of the “raw” model output (e.g., membrane voltage traces), but can also calculate the uncertainty and sensitivity of salient model response features such as spike timing, action potential width, average interspike interval, and other features relevant for various neural and neural network models. Uncertainpy comes with several common models and features built in, and including custom models and new features is easy. The aim of the current paper is to present Uncertainpy to the neuroscience community in a user-oriented manner. To demonstrate its broad applicability, we perform an uncertainty quantification and sensitivity analysis of three case studies relevant for neuroscience: the original Hodgkin-Huxley point-neuron model for action potential generation, a multi-compartmental model of a thalamic interneuron implemented in the NEURON simulator, and a sparsely connected recurrent network model implemented in the NEST simulator.

How Well Can Saliency Models Predict Fixation Selection in Scenes Beyond Central Bias? A New Approach to Model Evaluation Using Generalized Linear Mixed Models.

Since the turn of the millennium, a large number of computational models of visual salience have been put forward. How best to evaluate a given model's ability to predict where human observers fixate in images of real-world scenes remains an open research question. Assessing the role of spatial biases is a challenging issue; this is particularly true when we consider the tendency for high-salience items to appear in the image center, combined with a tendency to look straight ahead (“central bias”). This problem is further exacerbated in the context of model comparisons, because some—but not all—models implicitly or explicitly incorporate a center preference to improve performance. To address this and other issues, we propose to combine a-priori parcellation of scenes with generalized linear mixed models (GLMM), building upon previous work. With this method, we can explicitly model the central bias of fixation by including a central-bias predictor in the GLMM. A second predictor captures how well the saliency model predicts human fixations, above and beyond the central bias. By-subject and by-item random effects account for individual differences and differences across scene items, respectively. Moreover, we can directly assess whether a given saliency model performs significantly better than others. In this article, we describe the data processing steps required by our analysis approach. In addition, we demonstrate the GLMM analyses by evaluating the performance of different saliency models on a new eye-tracking corpus. To facilitate the application of our method, we make the open-source Python toolbox “GridFix” available.