Code Ocean

United States of America

Hierarchical Cell-based Feature Similarity Measurement

Bo Ma
Hierarchical cell-based similarity method efficiently measures the similarity of isosurfaces (1D) as well as isovalue-gradient magnitude features (2D) extracted from 3D volumetric datasets. This similarity measurement is used to integrate spatial information into the design of transfer functions in direct volume rendering as presented in "Volumetric Feature-Based Classification and Visibility Analysis for Transfer Function Design".

Hypercube Clustering

Oscar LiJen Hsu
Conventional sub-trajectory clustering is used to identify similarities among multiple trajectories. This method has been applied to traffic routes in urban areas as well as hurricane behavior. Existing methods tend to overlook many relevant sub-trajectories; others require a road network as input; all are slowed down considerably by large datasets. In this paper, we design a novel approach to sub-trajectory clustering in which trajectories are transformed into a set of hypercubes with four dimensions. The...

LPLDA: Local Pairwise Linear Discriminant Analysis

Heliang
Linear discriminant analysis - probabilistic linear discriminant analysis (LDA-PLDA) is a standard and effective backend in the field of speaker verification. The object of LDA is to perform dimensionality reduction while minimizing within-class covariance and maximizing between-class covariance. For a target class (or speaker), our task is to make a binary decision about whether a test utterance is from a specific target speaker. Generally, the non-target test utterances which are close to the target speaker...

Contact Hypothesis Reevaluated

Seth Green
This code reproduces the statistical analyses for "The Contact Hypothesis Revisited", by Betsy Levy Paluck, Seth A. Green and Donald P. Green.For a preprint, see https://osf.io/preprints/socarxiv/w2jkf. This paper evaluates the state of contact hypothesis research from a policy perspective. Building on Pettigrew and Tropp’s (2006) influential meta-analysis, we assemble all intergroup contact studies that feature random assignment and delayed outcome measures, of which there are 27 in total, nearly two-thirds of which were published following...

Enhanced PUMA for direction-of-arrival estimation and its performance analysis

Cheng
Direction-of-arrival (DOA) estimation is a problem of significance in many applications. In practice, due to the occurrence of coherent signals and/or when the number of available snapshots is small, it is a challenge to find DOAs accurately. This problem is revisited here through a new enhanced principal-singular-vector utilization for modal analysis (EPUMA) DOA estimation approach, which improves the threshold performance by first generating ($P+K$) DOA candidates for $K$ sources where $P\geq K$, and then judiciously...

A Global Optimal Solution to the Eco-Driving Problem - Hybrid Electric Heavy-Duty Vehicle

In this example, a global optimal solution to the eco-driving problem with velocity contraints for hybrid electric heavy-duty vehicle is compared to a constant speed profile driving.

A Global Optimal Solution to the Eco-Driving Problem - Benchmark Problem for Electric Vehicles.

In this example, we revisit the numerical benchmark problem for eco-driving that have been introduced in [10] to provide a global optimal solution using a SQP approach.

Linear Filtering Approximation Functions Comparison

Mariano
In this script 4 approximation functions (Butterworth, Chebyshev, Bessel and Cauer) for filtering applications are compared for several polynomial orders. As a result, the magnitude, phase and group delay, frequency response are plotted, together with the poles and zeros diagram of each filter.

Bi-level evolutionary multi-objective optimization of Echo State Network Autoencoder

Naima Chouikhi
Echo State Network is a recurrent randomized architecture in which the hidden states are implemented by a non-trained recurrent hidden layer. ESN is used as a recurrent Autoencoder (ESN-RAE) where the inputs are equal to the outputs. Both of basic and Multi-layer ESNs are used as Autoencoders. ESN (basic and ML) has some limits especially related to the setting of its architecture and some weights parameters. Providing an ”optimal” reservoir(s) for a given problem is...

Alejandro Garces
The power flow is a non-linear problem that requires a Newton's method to be solved in dc microgrids with constant power terminals. This paper presents sufficient conditions for the quadratic convergence of the Newton's method in this type of grids. The classic Newton method as well as an approximated Newton Method are analyzed in both master-slave and island operation with droop controls. Requirements for the convergence as well as for the existence and uniqueness of...

A Linear Three-Phase Load Flow for Power Distribution Systems

Alejandro Garces
This letter proposes a linear load flow for three-phase power distribution systems. Balanced and unbalanced operation are considered as well as the ZIP models of the loads. The methodology does not require any assumption related to the R/X ratio. Despite its simplicity, it is very accurate compared to the conventional back-forward sweep algorithm.

Demonstration of Discrimination Index (DI) Proposed for Fault and Power Swing Detection in Distance Relays

İbrahim Gürsu Tekdemir
Discrimination Index (DI) proposed in the article is demonstrated in this capsule. There are four cases simulated for that purpose. They are: 1) a 5-cycle synthetically created sine wave (a simple abnormal/unhealthy condition detection test) and three cases simulated for Single Machine - Infinite Bus (SMIB) test system: 2) short circuit fault, 3) power swing and 4) short circuit fault during power swing cases. In the first case (the simple abnormal/unhealthy condition detection test), amplitude...

Particle swarm optimization of Echo State Network parameters

Naima Chouikhi
ESN is a simple and powerful network. It is simple thanks to its non-complex architecture as well as its training method. It is powerful thanks to the good results given in the field of machine learning. Moreover, It has a special topology characterized by random parameters initialization especially those related to the reservoir and the weights. Although this random initialization is followed by some pre-treatments such as the scaling of the reservoir matrix by its...

Echo State Network Autoencoder ESN-AE

Naima Chouikhi
The presented code consists of an Echo State Network Recurrent Autoencoder ESN-RAE. In fact, sometimes the original data representations may not be the most expressive data distribution for a targeted task. The idea behind using AEs is to make the inputs equal to the outputs within a neural network then take the activations of the hidden layer as new data representation. ESN is used in this code as RAE to extract new data different from...

LPANNI: Label Propagation Algorithm with Neighbor Node Influence

Meilian Lu
LPANNI (Label Propagation Algorithm with Neighbor Node Influence) is an improved overlapping community detection algorithm, which detects overlapping community structures by adopting fixed label propagation sequence based on the ascending order of node importance and label update strategy based on neighbor node influence and historical label preferred strategy. The community package contains the algorithm source code, the details are as follows: countPath：Calculate all paths between a pair of nodes in a network and store them...

Who were the voters behind the Schulz effect? A long-term analysis of individual voter trajectories in the run-up to the 2017 German federal election (Wuttke/Schoen)

Alexander Wuttke
In early 2017, after the nomination of Martin Schulz as candidate for chancellor the SPD experienced a rapid surge in public support as measured in public opinion polls. For a short period the SPD seemed capable of becoming the largest party in the 2017 German federal election. Yet, the upward trend proved short-lived and the SPD ended up with the worst election result since 1949. Using data from a multi-wave panel survey, this analysis demonstrated...

XPlode: Explaining Repaired Data with CFDs

Joeri Rammelaere
XPlode is an algorithm for explaining an observed set of modifications by a user, with the goal of cleaning the data. The explanations take the form of a Conditional Functional Dependency. This source code accompanies the paper "Explaining Repaired Data with CFDs": *Abstract* Many popular data cleaning approaches are rule-based: Constraints are formulated in a logical framework, and data is considered dirty if constraints are violated.These constraints are often discovered from data, but to ascertain...

Bi-level evolutionary multi-objective optimization of Echo State Network Autoencoder

Naima Chouikhi
Echo State Network is a recurrent randomized architecture in which the hidden states are implemented by a non-trained recurrent hidden layer. ESN is used as a recurrent Autoencoder (ESN-RAE) where the inputs are equal to the outputs. Both of basic and Multi-layer ESNs are used as Autoencoders. ESN (basic and ML) has some limits especially related to the setting of its architecture and some weights parameters. Providing an ”optimal” reservoir(s) for a given problem is...

Phase Congruency Parameter Optimization

Mahdi Alavi
Inspired by the work of Oppenheim and Lim (IEEE Proceedings, vol. 69, pp. 529–541, 1981) who demonstrated that the phase of the image contains more information than its magnitude, I have been developing novel tools for the optimal detection of image features based on the concept of phase congruency (PC). PC principle states that image features are mainly perceived at the points where the Fourier components of the image are maximally in phase. Many papers...

ANNOgesic - A Swiss army knife for the RNA-Seq based annotation of bacterial/archaeal genomes

Sung-Huan Yu
ANNOgesic is the swiss army knife for RNA-Seq based annotation of bacterial/archaeal genomes. It is a modular, command-line tool that can integrate different types of RNA-Seq data based on dRNA-Seq (differential RNA-Seq) or RNA-Seq protocols that inclusde transcript fragmentation to generate high quality genome annotations. It can detect genes, CDSs/tRNAs/rRNAs, transcription starting sites (TSS) and processing sites, transcripts, terminators, untranslated regions (UTR) as well as small RNAs (sRNA), small open reading frames (sORF), circular RNAs,...

Interactive demonstration of plateau-generating mechanisms in temporal order judgments (TOJs)

Jan Tünnermann
This capsule provides an interactive demonstration of plateau-generating mechanisms in temporal order judgments (TOJs) to accompany our publication "Stuck on a Plateau? A Model-Based Approach to Fundamental Issues in Visual Temporal-Order Judgments". The TVA-reset model assumes a processing reset when the second stimulus is presented in brief succession after the first stimulus (assumed to equal TVA's t0 parameter). The alternative range of indecision model was proposed by García-Pérez & Alcalá-Quintana.

Single- and Multi-Objective Particle Swarm Optimization of Echo State Network architecture

Naima Chouikhi
Echo State Networks ESNs are specific kind of recurrent networks providing a black box modeling of dynamic non-linear problems. Their architecture is distinguished by a randomly recurrent hidden infra-structure called dynamic reservoir. Coming up with an efficient reservoir structure depends mainly on selecting the right parameters including the number of neurons and connectivity rate within it. Despite expertise and repeatedly tests, the optimal reservoir topology is hard to be determined in advance. Topology evolving can...

System for Statistical Contextual Structure of Representation for Game Studies

Rodrigo De Godoy Domingues
This project was design, and is in constant evolution, in order to provide a method to capture data in a structured manner that provides comparative operations and statistical manipulations with the intent of better understand games, their structure and motif.

Particle swarm optimization of Echo State Network parameters

Naima Chouikhi
ESN is a simple and powerful network. It is simple thanks to its non-complex architecture as well as its training method. It is powerful thanks to the good results given in the field of machine learning. Moreover, It has a special topology characterized by random parameters initialization especially those related to the reservoir and the weights. Although this random initialization is followed by some pre-treatments such as the scaling of the reservoir matrix by its...

Compressive Array Feed Network Design Algorithm

Heinrich Edgar Arnold Laue
Compressive arrays combine the signals from multiple antenna elements for a reduced number of beamforming controls. This algorithm is able to minimise the sidelobe level (SLL) of a compressive array for a given main-beam region and is able to exploit a limited scanning range for increased directivity. Constraints may also be placed on the subarray patterns outside the scanning range, for example by the placement of a hard null for the suppression of interference before...

• Software
344

• 2018
178
• 2017
140
• 2016
26

• 2018
178
• 2017
166

• Code Ocean
344