Find research datasets worth reusing

opencc-by-3.0-usMay 2022View details →

Network-Nanostructured ZIF-8 too Enable Percolation for Enhanced Gas Transport (Data Repository)

Data repository

opencc-by-4.0May 2022View details →

NCBI database for Lysobacteraceae network of gene sharing study

NCBI database for Lysobacteraceae network of gene sharing study

opencc-by-4.0May 2022View details →

Binding networks identify targetable protein pockets for mechanism-based drug design

This repository contains supporting files for the manuscript entitled: Binding networks identify targetable protein pockets for mechanism-based drug design.

A Probabilistic Framework for Mutation Testing in Deep Neural Networks - Models archive Part 2

Models used as part of the paper "A Probabilistic Framework for Mutation Testing in Deep Neural Networks ?" submitted to the journal Information and Software Technology Replication package using the data is available at https://github.com/FlowSs/PM

A Probabilistic Framework for Mutation Testing in Deep Neural Networks - Models archive Part 3

A Probabilistic Framework for Mutation Testing in Deep Neural Networks - Models archive Part 1

opencc-by-4.0May 2022View details →

Bayesian Network analysis for Single Cell Multiomics

The data from stratified random samples of scRNA expression, surface marker and SNF cluster membership were integrated with high-resolution CT (HRCT) Scores of COVID-19 patients. The healthy and recovered individuals were assigned an HRCT score of zero, indicating absence of active pneumonia. The integrative modeling analysis was carried out using the wiseR package for end-to-end Bayesian network learning, inference and dashboard deployment. All continuous variables in the integrated data were discretized using the k-means algorithm with k=3 for biological interpretability as low, medium and high. A discrete Bayesian Network was learned from the data using hill climbing optimization for finding the directed acyclic graph encoding the structural dependencies between variables. Eleven Bayesian network structures were ensembled averaged to derive the consensus structure. The consensus structure was then parametrized with marginal and conditional probability distributions using Monte Carlo Markov Chain (MCMC) approximate inference method.

opencc-zeroJun 2022View details →

Data from: A stochastic generative model for citation networks among academic papers

We propose a stochastic generative model to represent a directed graph constructed by citations among academic papers, where nodes and directed edges represent papers with discrete publication time and citations respectively. The proposed model assumes that a citation between two papers occurs with a probability based on the type of the citing paper, the importance of cited paper, and the difference between their publication times, like the existing models. We consider the out-degrees of citing paper as its type, because, for example, survey paper cites many papers. We approximate the importance of a cited paper by its in-degrees. In our model, we adopt three functions: a logistic function for illustrating the numbers of papers published in discrete time, an inverse Gaussian probability distribution function to express the aging effect based on the difference between publication times, and an exponential distribution (or a generalized Pareto distribution) for describing the out-degree distribution. We consider that our model is a more reasonable and appropriate stochastic model than other existing models and can perform complete simulations without using original data. In this paper, we first use the Web of Science database and see the features used in our model. By using the proposed model, we can generate simulated graphs and demonstrate that they are similar to the original data concerning the in- and out-degree distributions, and node triangle participation. In addition, we analyze two other citation networks derived from physics papers in the arXiv database and verify the effectiveness of the model.

opencc-by-4.0Jan 2022View details →

Predicting dry matter intake in Canadian Holstein dairy cattle using milk mid-infrared reflectance spectroscopy and other commonly available predictors via artificial neural networks

Supplementary Tables

Model Zoo: A Dataset of Diverse Populations of Neural Network Models - STL10 - Preprocessed Datasets

Abstract In the last years, neural networks have evolved from laboratory environments to the state-of-the-art for many real-world problems. Our hypothesis is that neural network models (i.e., their weights and biases) evolve on unique, smooth trajectories in weight space during training. Following, a population of such neural network models (refereed to as “model zoo”) would form topological structures in weight space. We think that the geometry, curvature and smoothness of these structures contain information about the state of training and can be reveal latent properties of individual models. With such zoos, one could investigate novel approaches for (i) model analysis, (ii) discover unknown learning dynamics, (iii) learn rich representations of such populations, or (iv) exploit the model zoos for generative modelling of neural network weights and biases. Unfortunately, the lack of standardized model zoos and available benchmarks significantly increases the friction for further research about populations of neural networks. With this work, we publish a novel dataset of model zoos containing systematically generated and diverse populations of neural network models for further research. In total the proposed model zoo dataset is based on six image datasets, consist of 24 model zoos with varying hyperparameter combinations are generated and includes 47’360 unique neural network models resulting in over 2’415’360 collected model states. Additionally, to the model zoo data we provide an in-depth analysis of the zoos and provide benchmarks for multiple downstream tasks as mentioned before. Dataset This dataset is part of a larger collection of model zoos and contains the zoos trained on the labelled samples from STL10. All zoos with extensive information and code can be found at www.modelzoos.cc. This repository contains the preprocessed model zoos wrapped in a custom pytorch dataset class (filenames beginning with "dataset"). Zoos are trained with small and large CNN models, in three configurations varying the seed only (seed), varying hyperparameters with fixed seeds (hyp_fix) or varying hyperparameters with random seeds (hyp_rand). Due to the large filesize, the raw datasets are hosted in a separate repository. The index_dict.json files contain information on how to read the vectorized models. For more information on the zoos and code to access and use the zoos, please see www.modelzoos.cc.

Model Zoo: A Dataset of Diverse Populations of Neural Network Models - USPS

Abstract In the last years, neural networks have evolved from laboratory environments to the state-of-the-art for many real-world problems. Our hypothesis is that neural network models (i.e., their weights and biases) evolve on unique, smooth trajectories in weight space during training. Following, a population of such neural network models (refereed to as “model zoo”) would form topological structures in weight space. We think that the geometry, curvature and smoothness of these structures contain information about the state of training and can be reveal latent properties of individual models. With such zoos, one could investigate novel approaches for (i) model analysis, (ii) discover unknown learning dynamics, (iii) learn rich representations of such populations, or (iv) exploit the model zoos for generative modelling of neural network weights and biases. Unfortunately, the lack of standardized model zoos and available benchmarks significantly increases the friction for further research about populations of neural networks. With this work, we publish a novel dataset of model zoos containing systematically generated and diverse populations of neural network models for further research. In total the proposed model zoo dataset is based on six image datasets, consist of 24 model zoos with varying hyperparameter combinations are generated and includes 47’360 unique neural network models resulting in over 2’415’360 collected model states. Additionally, to the model zoo data we provide an in-depth analysis of the zoos and provide benchmarks for multiple downstream tasks as mentioned before. Dataset This dataset is part of a larger collection of model zoos and contains the zoos trained on the labelled samples from USPS. All zoos with extensive information and code can be found at www.modelzoos.cc. This repository contains two types of files: the raw model zoos as collections of models (file names beginning with "usps_"), as well as preprocessed model zoos wrapped in a custom pytorch dataset class (filenames beginning with "dataset"). Zoos are trained in three configurations varying the seed only (seed), varying hyperparameters with fixed seeds (hyp_fix) or varying hyperparameters with random seeds (hyp_rand). The index_dict.json files contain information on how to read the vectorized models. For more information on the zoos and code to access and use the zoos, please see www.modelzoos.cc.

Dataset: Gauze detection and segmentation in minimally invasive surgery video using convolutional neural networks

Dataset of the Gauze detection and segmentation in minimally invasive surgery video using convolutional neural networks article. Further information is available in the README file.

opencc-by-4.0Nov 2021View details →

PINNup: Robust neural network wavefield solutions using frequency upscaling and neuron splitting

Solving for the frequency-domain scattered wavefield via physics-informed neural network (PINN) has great potential in increasing the flexibility and reducing the computational cost of seismic modeling and inversion. We propose a novel implementation of PINN using frequency upscaling and neuron splitting, which allows the neural network model to grow in size as we increase the frequency while leveraging the information from the pre-trained model for lower-frequency wavefields, resulting in fast convergence to high-accuracy wavefield solutions. In this letter, we present the relevant dataset to the paper.

Ice water path retrievals from Meteosat-9 with quantile regression neural networks: video supplement

Supplementary videos used from in A. Amell, P. Eriksson, S. Pfreundschuh: Ice water path retrievals form Meteosat-9 with quantile regression neural networks.

Observation of Aharonov-Bohm effect in PbTe nanowire networks

This repository contains the raw data and processing Python scripts corresponding to the paper "Observation of Aharonov-Bohm effect in PbTe nanowire networks"

opencc-zeroJun 2022View details →

Western Diamondback Rattlesnake interaction matrices for network analysis

Social network ecology is a powerful framework to assess patterns of interconnectedness and identify group-level interactions. We investigated social network structure in a pitviper (Crotalus atrox) to determine if group-level interactions result in network structures for denning, pairing, and parentage. We tested if network centrality was influenced by body length, sex, home range size/location, or generic relatedness. We revealed that networks were structurally modular but not nested. Sex was the only significant predictor of centrality in the parentage network, likely due to high levels of multiple paternity. Genotypic data revealed interacting focal individuals were unlikely to be related in networks; however, analysis of a larger group of subjects suggests kin-association at communal dens. Den selection may be driven by a combination of social preference, experience, and/or genetic relatedness. We demonstrated strong fission-fusion dynamics connected to annual migrations to summer home ranges and use of communal winter dens. Furthermore, both sexes show high fidelity to home ranges and dens, but females occasionally alter den sites, indicating active manipulation of their social environment. Our study illustrates that comprehensive, long-term datasets incorporating social network analysis with spatial and genetic information provide robust and unique insights to understanding social structure of understudied, cryptic taxa.

opencc-zeroJun 2022View details →

Operational assimilation of spectral wave data from the Sofar Spotter network

Historically, the sparseness of in situ open-ocean wave and weather observations has severely limited the forecast skill of weather over the ocean with major social and economic consequences for coastal communities and maritime industries. Ocean surface waves, specifically, are important for the interaction between atmosphere and ocean, and thus key in modeling weather and climate processes. Here, we investigate the improvements achievable from a large distributed sensor network combined with advances in assimilation strategies. Wave spectra from a global network of over 600 Sofar Spotter buoys are assimilated into an operational global wave forecast via optimal interpolation to update model spectra to best fit observations. We demonstrate end-to-end improvements in forecast skill of significant wave height of 38%, and up to 45% for other bulk parameters. This shows distributed observations of the air-sea interface, with advances in assimilation strategies, can reduce uncertainty in forecasts to dramatically improve earth system modeling

opencc-by-sa-4.0Apr 2022View details →

Impact of training set size on the ability of deep neural networks to deal with label noise

This is the data set accompanying the publication 'Impact of training set size on the ability of deep neural networks to deal with label noise'   License This work is shared under the <code>Creative Commons Attribution ShareAlike 4.0 International License</code><code> (CC BY-SA 4.0)</code>: https://creativecommons.org/licenses/by-sa/4.0/ The dataset is taken from the 6th SpaceNet challenge (https://spacenet.ai/sn6-challenge/) which also falls under the <code>CC BY-SA 4.0 license</code>. The original dataset was modified by removing objects from the buildings json files and converting the json files to png images. Creators of the SpaceNet dataset: Shermeyer, J., Hogan, D., Brown, J., Etten, A.V., Weir, N., Pacifici, F., Hänsch, R., Bastidas, A., Soenen, S., Bacastow, T.M., & Lewis, R. SpaceNet partners: https://spacenet.ai/about-us/

Minimizing Search Areas for Leak Detection in Water Distribution Networks - Code

This database includes the code used to analyze and produce the results for the following research article: Minimizing Search Areas for Leak Detection in Water Distribution Networks by B. Snider, G. Lewis, A.S. Chen, L. Vamvakeridou-Lyroudia, S. Djordjevic, D.A. Savic. Journal of Hydroinformatics. (Accepted - awaiting publication).