Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
117
datasets available to search
ShareScore release 0.9.0
Dataset results
117 results for “Machine-learning”
Research data for "Exploring the configurational space of amorphous graphene with machine-learned atomic energies"
<p>This dataset supports the paper: "Exploring the configurational space of amorphous graphene with machine-learned atomic energies" (<a href="https://doi.org/10.1039/D2SC04326B">https://doi.org/10.1039/D2SC04326B</a>).</p> <p>Trajectory data for the 200-atom structures (Fig. 3) and the final configurations for the 612-atom structures as well as the GAP-17-optimised 610-atom structure from Toh et al are provided (Fig. 4). Additionally, the structures used for data analysis in Fig. 5 are given.</p> <p>The files are in extended xyz (.xyz) format and contain the raw data for coordinates, forces, and atomic energies (labelled 'c_1'). The files also contain the atomic energies relative to pristine graphene, labelled "Energy_per_atom", and the locally averaged energy relative to pristine graphene, labelled "NN_Energy_per_atom". Topological information is included at the end of the .xyz file for the 612-atom structures ('fig_4'/) and for the structures in 'fig_5/'.</p> <p>All raw atomic energies were computed using LAMMPS default settings and were output with six significant figures, with the exception of the Toh et al. structure (for which ASE was used, outputting a higher number of significant figures). </p> <p>The data can be read using, for example, the Atomic Simulation Environment (ASE), or visualised using Ovito.</p> <p> </p>
Data from: Performance of unmarked abundance models with data from machine-learning classification of passive acoustic recordings
<p>The ability to conduct cost-effective wildlife monitoring at scale is rapidly increasing due to availability of inexpensive autonomous recording units (ARUs) and automated species recognition, presenting a variety of advantages over human-based surveys. However, estimating abundance with such data collection techniques remains challenging because most abundance models require data that are difficult for low-cost monoaural ARUs to gather (e.g., counts of individuals, distance to individuals), especially when using the output of automated species recognition. Statistical models that do not require counting or measuring distances to target individuals in combination with low-cost ARUs provide a promising way of obtaining abundance estimates for large-scale wildlife monitoring projects but remain untested. We present a case study using avian field data collected in forests of Pennsylvania during the Spring of 2020 and 2021 using both traditional point counts and passive acoustic monitoring at the same locations. We tested the ability of the Royle-Nichols and time-to-detection models to estimate abundance of two species from detection histories generated by applying a machine-learning classifier to ARU-gathered data. We compared abundance estimates from these models to estimates from the same models fit using point-count data and to two additional models appropriate for point counts, the N-mixture model and distance models. We found that the Royle-Nichols and time-to-detection models can be used with ARU data to produce abundance estimates similar to those generated by a point-count based study but with greater precision. ARU-based models produced confidence or credible intervals that were on average 31.9% ( 11.9 SE) smaller than their point-count counterpart. Our findings were consistent across two species with differing relative abundance and habitat use patterns. The higher precision of models fit using ARU data is likely due to higher cumulative detection probability, which itself may be the result of greater survey effort using ARUs and machine-learning classifiers to sample significantly more time for focal species at any given point. Our results provide preliminary support the use of ARUs in abundance-based study applications, and thus may afford researchers a better understanding of habitat quality and population trends, while allowing them to make more informed conservation actions and recommendations.</p>
Hyperparameter tuning and performance assessment of statistical and machine-learning models using spatial data.
<p>This is a research compendium (RC) for the publication "Hyperparameter tuning and performance assessment of statistical and machine-learning algorithms using spatial data".</p> <p>The code (including figures, appendices and the manuscript) is packed in <strong>pathogen-modeling-3.zip </strong>or can be found directly in the <a href="https://github.com/pat-s/pathogen-modeling">Github repository</a>.</p> <ul> <li><strong>Publication figures</strong>: analysis/paper/submission/3/latex-source-files/</li> <li><strong>Appendices</strong>: analysis/paper/submission/3/</li> </ul> <p>This RC represents a static snapshot at the time of submission. The Github repository will receive changes after the publication was published.</p> <p><strong>Data sources</strong></p> <ul> <li>Atlas Climatico: <a href="http://opengis.uab.es/wms/iberia/index.htm">http://opengis.uab.es/wms/iberia/index.htm</a></li> <li>DEM: ftp://ftp.geo.euskadi.eus/lidar/MDE_LIDAR_2016_ETRS89/</li> <li>Lithology: <a href="http://www.geo.euskadi.eus/geonetwork/srv/spa/main.home">http://www.geo.euskadi.eus/geonetwork/srv/spa/main.home</a></li> <li>pH: <a href="https://esdac.jrc.ec.europa.eu/content/soil-ph-europe#tabs-0-description=0">https://esdac.jrc.ec.europa.eu/content/soil-ph-europe#tabs-0-description=0</a></li> <li>soil: <a href="https://www.isric.org/explore/soilgrids">https://www.isric.org/explore/soilgrids</a></li> </ul> <p><strong>Licenses</strong></p> <p>All files are shared via the given license with the exception of "soil.tif" which is shared via the <strong>ODbL </strong>license<strong>.</strong></p>
Molecular dynamics simulation of tricaproin in gas phase using machine-learning potential ANI2x
<p>Tricaproin (Glycerol trihexanoate) is an example of a triglyceride molecule with very short alkyl tails attached to the glycerol moiety, and this deposit contains a 10 ns long simulation of tricaproin in a gas phase.</p> <p>The model chemistry (a.k.a. interaction potential or force field) is the machine-learning potential ANI2x implemented in python package torchANI, which has a close-to-DFT accuracy, yet low cost compared to DFT or other electronic structure theories.</p> <p>Molecular dynamics were run using ASE with Langevin integrator at a constant temperature of 310 K.</p> <p>The resulting trajectory was written every 1 ps ("traj.h5", can be viewed in ASEgui), and gathered every 10 ps in a XTC format ("traj.xtc", open in MDAnalysis, VMD, UnityMol, GROMACS tools ...).</p> <p> </p> <p>Detailed simulation settings are in the python script.</p> <p> </p> <p>This simulation was performed for the purpose of building a coarse grained Martini 3 model of this molecule.</p> <p> </p> <p>ANI2x https://doi.org/10.26434/chemrxiv.11819268.v1</p> <p>torchANI https://aiqm.github.io/torchani/index.html</p> <p>ASE https://wiki.fysik.dtu.dk/ase/tutorials/md/md.html#constant-temperature-md</p>
A Supervised Machine-Learning Prediction of Textile's Antimicrobial Capacity Coated with Nanomaterials
<p>The dataset contains P-Chem properties of NMs and experimental conditions for assessing the antimicrobial properties of inorganic and organic NMs using machine learning tools.</p>
Supporting Information for the Journal Article "Quantum Chemical Data Generation as Fill-In for Reliability Enhancement of Machine-Learning Reaction and Retrosynthesis Planning"
<p>This data set contains all data produced when exploring the Williamson ether synthesis starting from iodoethane and phenol.</p> <p><br> The set is structures as follows:</p> <ul> <li>analysis: Contains the script used to analyze the exploration and the output of said script</li> <li>check_barrier: Contains the output of the manual calculations done to check the barrier of the reaction</li> <li>exploration: Contains the scripts used to initialize and carry out the exploration as well as the two starting structures as XYZ files</li> <li>raw_data: a dump of the MongoDB database with all the data produced during the exploration</li> </ul>
CCClim - A machine-learning powered cloud class climatology
<p>CCClim is based on cloud property retrievals from the European Space Agency's (ESA) Cloud\_cci dataset, adding relative occurrences of eight major cloud types as defined by the World Meteorological Organization (WMO) at 1° resolution. </p><p>The cloud types are predicted using a two stage machine learning framework, in which a 1 km pixel-level classifier is followed up with a grid-box scale Random Forest regression model.</p><p>CCClim's global coverage being almost gapless from 1982 to 2016 allows for performing process-oriented analyses of clouds on a climatological time scale. Similarly, the moderate spatial and temporal resolutions make it a lightweight dataset while enabling straightforward comparison to climate models.</p><p>The compressed tarball contains 35 netCDF files, each covering one calendar year. Each file provides daily averages of nine cloud-related variables and the nine classes (eight cloud types+undetermined) as per 1° grid box fractional amounts.</p><p>Cloud-related variables:</p><ul><li>cloud water path</li><li>ice water path</li><li>liquid water path</li><li>cloud optical depth</li><li>effective liquid droplet radius at cloud top</li><li>effective ice particle radius at cloud top</li><li>cloud top pressure</li><li>surface temperature</li><li>cloud area fraction</li></ul><p>Cloud types:</p><ul><li>Ci: Cirrus/Cirrostratus</li><li>As: Altostratus</li><li>Ac: Altocumulus</li><li>St: Stratus</li><li>Sc: Stratocumulus</li><li>Cu: Cumulus</li><li>Ns: Nimbostratus</li><li>Dc: Deep convective</li></ul>
Bee Tracker – an open-source machine-learning based video analysis software for the assessment of nesting and foraging performance of cavity-nesting solitary bees
Open the record for dataset details and reuse information.
Data from: Performance of unmarked abundance models with data from machine-learning classification of passive acoustic recordings
Open the record for dataset details and reuse information.
Dataset for "Mo-Si alloys studied by atomistic computer simulations using a novel machine-learning interatomic potential: Thermodynamics and interface phenomena"
<p>This dataset was used to fit a general purpose machine-learning interatomic potential for Mo-Si alloys based on the Atomic Cluster Expansion (ACE) formalism. It supports the paper "Mo-Si alloys studied by atomistic computer simulations using a novel machine-learning interatomic potential: Thermodynamics and interface phenomena".</p>
Machine-learning Generated Catalog of Long Period Variables from Palomar Gattini-Infrared Lightcurves
<pre>A catalog of LPVs obtained from a decision tree classifier trained on features extracted from Palomar Gattini-IR J band <br>lightcurves as described in Suresh et. al. 2024 (https://arxiv.org/abs/2402.08000).<br><br>Description of columns<br><br>1. 'Name' - Unique identifier for each LPV; internal designation of the object<br>2. 'RA' - Right ascension<br>3. 'Dec' - Declination<br>4. 'num_detections' - number of detections<br><br>Calculated features:<br>5. 'von_neumann_score' - von-Neumann score <br>6. 'gp_score' - Gaussian process regression fit score<br>7. 'slope_min' - minimum slope of the lightcurve<br>8. 'slope_max' - maximum slope of the lightcurve<br>9. 'ptp' - peak-to-peak amplitude <br>10. 'nflips' - number of flips<br>11. 'max_rate' - maximum point-wise slope <br>12. 'min_rate' - minimum point-wise slope 13. 'lcdur' - observation baseline<br>14. 'per_diff_90_50' - difference between 90th and 50th percentiles of J band magnitude<br>15. 'per_diff_95_50' - difference between 95th and 50th percentiles of J band magnitude <br>16. 'J' - Stetson J index<br>17. 'K_s' - Stetson K index, calculated as single band data<br>18. 'K_m' - Stetson K index, calculated as multi-band data <br>19. 'L_s' - Stetson L index, calculated as single band data 20. 'L_m' - Stetson L index, calculated as multi-band data<br>21. 'phase_chisq' - chi sq of sinusiod fit to phase folded lightcurve<br>22. 'phase_redchi' - reduced chi sq of sinusiod fit to phase folded lightcurve <br>23. 'phase_aic' - akaike information criterion of sinusiod fit to phase folded lightcurve <br>24. 'phase_bic' - bayesian information criterion of sinusiod fit to phase folded lightcurve 25. 'bestperiod' - best fit Lomb-Scargle period <br>26. 'bestperiod_1' - second most likely Lomb-Scargle period<br>27. 'bestperiod_2' - third most likely Lomb-Scargle period<br>28. 'LSscore' - Lomb-Scargle score of best period <br>29. 'LSscore_2' - Lomb-Scargle score corresponding to second best period 30. 'LSscore_3' - Lomb-Scargle score corresponding to third best period <br>31. 'amplitude' - J band amplitude <br>32. 'meanF' - mean predicted value from Lomb-Scargle fit <br>33. 'chi2' - chi sq of sinusoidal fit to lightcurve<br>34. 'redchi2' - reduced chi square of sinusoidal fit to lightcurve<br>35. 'rednullchi2' - reduced chi square of linear fit to lightcurve 36. 'nullchi2' - chi square of linear fit to lightcurve<br>37. 'maxperiod' - maximum period from Lomb-Scargle fit <br>38. 'linear_slope' - slope of linear fit to lightcurve<br>39. 'linear_intercept' - y-intercept of linear fit to lightcurve<br>40. 'J_mag' - mean J band magnitude<br>41. 'LSratio' - ratio of Lomb-Scargle score of two most promiment peaks in <br> Lomb-Scargle periodogram<br>42. 'chi2ratio' - ratio of reduced chi square of sinusoidal fit and reduced <br> chi square of linear fit to lightcurve <br>43. 'period_ratio' - ratio of best fit Lomb-Scargle period and <br> maximum Lomb-Scargle period<br>44. 'prob_lpv_sum' - machine-learning classifier LPV score (obtained by summing probability of LPV and type II LPV)<br><br>Color information: 45. 'w1mpro' - WISE W1 mag<br>46. 'w1sigmpro' - WISE W1 error <br>47. 'w2mpro' - WISE W2 mag<br>48. 'w2sigmpro' - WISE W2 error <br>49. 'w3mpro' - WISE W3 mag<br>50. 'w3sigmpro' - WISE W3 error 51. 'w4mpro' - WISE W4 mag<br>52. 'w4sigmpro' - WISE W4 error<br>53. 'j_m' - 2MASS J mag <br>54. 'j_msigcom' - 2MASS J error<br>55. 'h_m' - 2MASS H mag <br>56. 'h_msigcom' - 2MASS H error<br>57. 'k_m' - 2MASS K mag 58. 'k_msigcom' - 2MASS K error<br>59. 'J-H' - 2MASS J-H<br>60. 'J-K' - 2MASS J-K 61. 'W1-W2' - WISE W1-W2 <br>62. 'W3-W4' - WISE W3-W4 <br>63. 'W1-W4' - WISE W1-W4</pre>
Ab initio data to generates machine-learned force fields of ions in aqueous medium in VASP format.
<p>ML_AB_H: Datasets of 64 water molecules and a single proton with 64 water molecules.</p> <p>ML_AB_VP2: Datasets of a single V^2+ ion with 64 water molecules.</p> <p>ML_AB_VP3: Datasets of a single V^3+ ion with 64 water molecules.</p> <p>ML_AB_FeP2: Datasets of a single Fe^2+ ion with 64 water molecules.</p> <p>ML_AB_FeP3: Datasets of a single Fe^3+ ion with 64 water molecules.</p> <p>ML_AB_CuP1: Datasets of a single Cu^+ ion with 64 water molecules.</p> <p>ML_AB_CuP2: Datasets of a single Cu^2+ ion with 64 water molecules.</p> <p>ML_AB_RuP2: Datasets of a single Ru^2+ ion with 64 water molecules.</p> <p>ML_AB_RuP3: Datasets of a single Ru^3+ ion with 64 water molecules.</p> <p>ML_AB_AgP1: Datasets of a single Ag^+ ion with 64 water molecules.</p> <p>ML_AB_AgP2: Datasets of a single Ag^2+ ion with 64 water molecules.</p> <p>ML_AB_O2: Datasets of a single O2 ion with 64 water molecules.</p> <p>ML_AB_O2N1: Datasets of a single O2^- ion with 64 water molecules.</p> <p>ML_AB_water: Datasets of 64 water molecules presenting bulk water and 64 water molecules representing water slab.</p> <p>All datasets were generated by VASP using PAW, plane wave basis sets with cutoff energy of 520 eV and RPBE+D3 exchange-correlation functional with zero-damping. All ab initio calculations were done on extended systems with periodic boundary conditions. See details in <a href="https://doi.org/10.48550/arXiv.2409.11000">https://doi.org/10.48550/arXiv.2409.11000</a>.</p>
Dataset for: Anniés et al., "Accessing structural, electronic, transport and mesoscale properties of Li-GICs via a complete DFTB-model with machine-learned repulsion potential"
<p>GPrep training data, GPrep jupyter notebook, .skf files.</p> <p>The GPrep code is available at https://doi.org/10.5281/zenodo.3697913</p>
WYC seismic phases for "Lighting up an 1-km fault near a hydraulic fracturing well using machine-learning based picker"
<p>In this study, we applied a state-of-the-art package on newly collected nodal-array data around a hydraulic-fracturing well. The array consists of up to 85 nodes with an average station spacing of less than a kilometer. Within the hydraulic-fracturing stimulation weeks, we detected ~3000 seismic events with magnitude down to ~-2. </p> <p>The seismic phases to associate the events are included in Zenodo_share.zip. The final catalog and station locations (in relative scale) are included in the excel spreadsheet.</p>
Machine-Learning Based Location of the 2021 MW 7.4 Maduo Earthquake Sequence: Insight into Intraplate Seismogenesis
<p>This file is the Machine-Learning Based earthquake catalog of the 2021 MW 7.4 Maduo Earthquake. It is only used for scientific research.</p> <p>ATTENTION!!!</p> <p>The data for the paper "Relocation of the 2024 MS 7.1 Wushi, Xinjiang earthquake sequence and implications for seismogenic structure" is accessible at the website "https://zenodo.org/records/12790377".</p> <p>ATTENTION!!!</p>
Data from: OoCount: A machine-learning based approach to mouse ovarian follicle counting and classification
<p>The number and distribution of ovarian follicles in each growth stage provides a reliable readout of ovarian health and function. Leveraging techniques for three-dimensional (3D) imaging of ovaries in toto has the potential to uncover total, accurate ovarian follicle counts. However, because of the size and holistic nature of these images, counting oocytes is time consuming and difficult. The advent of deep-learning algorithms has allowed for the rapid development of ultra-fast, automated methods to analyze microscopy images. In recent years, these pipelines have become more user-friendly and accessible to non-specialists. We used these tools to create OoCount, a high-throughput, open-source method for automatic oocyte segmentation and classification from fluorescent 3D microscopy images of whole mouse ovaries using a deep-learning convolutional neural network (CNN) based approach. We developed a fast clearing and spinning disk confocal-based imaging protocol to obtain 3D images of whole mount perinatal and adult mouse ovaries. Then, fluorescently labeled oocytes from 3D images of ovaries were manually annotated to develop a machine learning training dataset. This dataset was used to train a CNN to automatically label all oocytes in the ovary. In a second phase, we trained another CNN to classify labeled oocytes and sort them into growth stages. Using OoCount, we can obtain accurate counts of oocytes in each growth stage in the perinatal and adult ovary, improving our ability to study ovarian function and fertility. Here, we provide an end-to-end protocol for developing high quality 3D images of the perinatal and adult mouse ovary, obtaining follicle counts and stages, and how to customize OoCount to fit images produced in any lab.</p>
Trained Potentials for Article "Computationally Efficient Machine-Learned Model for GST Phase Change Materials via Direct and Indirect Learning"
<p>We provide 8 files here to get started using our trained potentials:</p> <p>1) *.yaml files for each trained potential. These are the outputs of the PACE training process.</p> <p>2) *.yace files for each trained potential. These are read by LAMMPS to use the trained potential. They can be obtained from the *.yaml files using the command line command: "pace_yaml2yace *.yaml".</p> <p>3) GST_config.data -- a starting configuration of GST to be read by LAMMPS. This configuration contains 504 atoms at density 5.85 g/cm^3.</p> <p>4) sample.inp -- a sample LAMMPS input file using the trained potentials. This currently uses "ACE-Indir2.yace" to run the starting configuration "GST_config.data" for 10 ps at 1200 K. When run, it outputs a log file "test.log" and a dump file "test.dump". The choice of trained potential can be changed in the "pair_coeff" section.</p>
Datatset: Machine-Learning Side-Channel Attacks on the GALACTICS Constant-Time Implementation of BLISS
<p>This dataset accompanies the paper "Machine-Learning Side-Channel Attacks on the GALACTICS Constant-Time Implementation of BLISS". It was used to experimentally prove the presented attack strategies on real hardware. The corresponding source code for all three attacks is also publicly available.</p> <p>A detailed description of how the data was obtained can be found in the paper. Section 4 precisely describes the experimental setup.</p> <p> </p> <p>Prerequisites:</p> <pre><code class="language-bash">sudo apt-get install p7zip</code></pre> <p> </p> <p>Extract the data:</p> <pre><code class="language-bash">7z x galactics_attack_data.7z</code></pre> <p> </p> <p>Running the attacks:</p> <p>The source code to run the three presented attacks can be found on Github. The instructions on how to use the python code can be obtained from the corresponding README.</p> <p> </p> <p>Re-using the dataset:</p> <p>The dataset consists of <em>.pickle</em> and <em>.bin</em> files. The <em>.pickle</em> files can be read using <a href="https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.read_pickle.html">Pythons Pandas library</a>. Python access functions for the <em>.bin</em> files are also provided.</p>
Consensus machine-learning models for protein-ligand binding affinity estimation
<p><strong>Motivation:</strong> In structure-based virtual screening, machine learning based scoring function gained popularity in the last few years as they outperformed classical scoring function. The protein-ligand system can be encoded by a set of orthogonal descriptor spaces, which are then mined by machine learning algorithms to find a relationship with the binding affinity experimental value.</p> <p><strong> </strong></p> <p><strong>Results:</strong> In this work we propose our modelling approach to derive a new scoring function, derived from a combination of multiple descriptor spaces coupled with machine learning algorithms ensembled in consensus. The SF has been trained on the PDBbind v.2019 data and has been extensively internally and externally validated on a large set of complexes. When benchmarked on the PDBbind core set, it achieved better performance than state-of-the-art counterparts, scoring: R<sub>Pearson </sub>= 0.85-0.86 r<sup>2</sup> = 0.70-0.72 and RMSE = 1.15-1.21. As highlights: (i) an applicability domain definition has been implemented to delimit the SF’s application boundaries, and (ii) a mechanistic interpretation is proposed by investigating the contribution of each protein-ligand atom pairs in the prediction of the binding affinity, which could provide a support in the lead-optimization process.</p> <p><strong> </strong></p> <p><strong>Availability and implementation:</strong> Our scoring function is freely available through the webportal: <a href="https://predictor.exscalate.eu/">https://predictor.exscalate.eu/</a></p>
TerraSenseTK - Towards Reproducible Machine-Learning and Remote Sensing Research
<p>Dataset used in TerraSenseTK - Towards Reproducible Machine-Learning and<br> Remote Sensing Research.</p> <p>Nutrient Estimation in Common wheat Case Study available in the notebook</p> <p>Documentation is available in <a href="https://terrasensetk.readthedocs.io/en/latest/">here</a>.</p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.