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.
1,943
datasets available to search
ShareScore release 0.9.0
Dataset results
1,943 results for “machine learning”
Machine Learning for Bird Song Learning (ML4BL) dataset
<p><strong>General description</strong></p> <p>This dataset contains Zebra Finch decisions about perceptual similarity on song units. All the data and files are used for reproducing the results of the paper 'Bird song comparison using deep learning trained from avian perceptual judgments' by the same authors. </p> <p><strong>Git repo on Zenodo:</strong> <a href="https://doi.org/10.5281/zenodo.5545932">https://doi.org/10.5281/zenodo.5545932</a><br> <strong>Git repo access: </strong><a href="https://github.com/veronicamorfi/ml4bl/tree/v1.0.0">https://github.com/veronicamorfi/ml4bl/tree/v1.0.0</a></p> <p><strong>Directory organisation:</strong><br> ML4BL_ZF<br> |_files<br> |_Final_probes_20200816.csv - all trials and decisions of the birds (aviary 1 cycle 1 data are removed from experiments)<br> |_luscinia_triplets_filtered.csv - triplets to use for training<br> |_mean_std_luscinia_pretraining.pckl - mean and std of luscinia triplets used for trianing<br> |_*_cons_* - % side consistency on triplets (train/test) - train set contains both train and val splits<br> |_*_gt_* - cycle accuracy for triplets of the specific bird (train/test) - train set contains both train and val splits<br> |_*_trials_* - number of decisions made for a triplet (train/test) - train set contains both train and val splits<br> |_*_triplets_* - triplet information (aviary_cycle-acc_birdID, POS, NEG, ANC) (train/test) - train set contains both train and val splits<br> |_*_low*_ - low-margin (ambiguous) triplets (train/val/test)<br> |_*_high_ - high-margin (unambiguous) triplets (train/val/test)<br> |_*_cycle_bird_keys_* - unique aviary_cycle-acc_birdID keys (train/test) - train set contains both train and val splits<br> |_TunedLusciniaV1e.csv - pairwise distance of two recordings computed by Luscinia<br> |_training_setup_1_ordered_acc_single_cons_50_70_trials.pckl - dictionary containing everything needed for training the model (keys: 'train_keys', 'train_triplets', 'val_keys', 'vali_triplets', 'test_triplets', 'test_keys', 'train_mean', 'train_std')<br> |_melspecs - *.pckl - melspectrograms of recordings<br> |_wavs - *wav - recordings<br> |_README.txt</p> <p><strong>Recordings</strong></p> <p>887 syllables extracted from zebra finch song recordings, with a sampling rate of 48kHz and high pass filtered (100Hz), with a 20ms intro/outro fade. </p> <p><strong>Decisions</strong></p> <p>Triplets were created from the recordings and the birds made side based decisions about their similarity (see 'Bird song comparison using deep learning trained from avian perceptual judgments' for further information).</p> <p><strong>Training dictionary Information</strong></p> <p>Dictionary keys:<br> 'train_keys', 'train_triplets', 'val_keys', 'vali_triplets', 'test_triplets', 'test_keys', 'train_mean', 'train_std'</p> <p>train_triplets/vali_triplets/test_triplets: <br> Aviary_Cycle_birdID, POS, NEG, ANC, Decisions, Cycle_ACC(%), Consistency(%)<br> <br> train_keys/val_keys/test_keys:<br> Aviary_Cycle_birdID</p> <p>train_mean/train_std:<br> shape: (1, mel_bins)</p> <p> </p> <p><strong>Open Access</strong></p> <p>This dataset is available under a Creative Commons Attribution 4.0 International (CC BY 4.0) license.</p> <p><br> <strong>Contact info</strong></p> <p>Please send any questions about the recordings to:<br> Lies Zandberg: Elisabeth.Zandberg@rhul.ac.uk</p> <p>Please send any feedback or questions about the code and the rest of the data to:<br> Veronica Morfi: g.v.morfi@qmul.ac.uk</p>
SNP genotype matrix for GWAS and Machine Learning analyses
<p><strong>SNP datasets used for GWAS and Machine Learning analyses</strong></p> <p>All datasets come from the easyGWAS website: <a href="https://easygwas.ethz.ch/down/1/">https://easygwas.ethz.ch/down/1/</a></p> <p> </p> <p><strong>=== Horton et al. 2012 ===</strong></p> <p><strong>1307 Arabidopsis genotypes x 214,057</strong> <strong>SNPs</strong></p> <p><strong>1) In the form of a genotype matrix </strong></p> <p>The file is called <a href="https://zenodo.org/api/files/d862e79f-02f2-4176-9b8e-04e48a2cf72c/horton2012.raw?versionId=2b05fb8d-b486-4f00-b7bf-bb024b892dc9">Horton2012.raw</a></p> <p><a href="https://www.nature.com/articles/ng.1042">https://www.nature.com/articles/ng.1042</a></p> <p>Preview of the first lines and columns:</p> <p>FID Chr1_657_T Chr1_3102_G Chr1_4648_A Chr1_4880_T Chr1_5975_G Chr1_6063_T Chr1_6449_C<br> 9381 2 2 2 0 0 0 0<br> 9380 0 0 0 0 0 0 2<br> 9378 2 2 2 0 0 0 0<br> 9371 2 2 2 0 0 0 0<br> 9367 0 0 0 2 0 0 0<br> 9363 2 2 2 0 0 0 0<br> 9356 0 2 2 0 0 0 0<br> 9355 2 2 2 0 0 0 0<br> 9354 2 2 2 0 0 0 0</p> <p>...etc...</p> <p>PLINK 1.9 was used to convert the .ped and .map file to a .raw format with: </p> <pre><code class="language-bash">plink --file original_data/genotype --recodeA --tab</code></pre> <p>Genotypes are encoded as 0, 1 or 2 with:</p> <pre> SNP SNP_A --- ----- A A -> 0 A C -> 1 C C -> 2 0 0 -> NA </pre> <p>Then only the Family ID was kept (same as individual ID) and other columns (Paternal ID, Maternal ID, Sex, Phenotype) were removed.</p> <p>The corresponding PLINK manual page used is here: <a href="https://zzz.bwh.harvard.edu/plink/dataman.shtml#recode">https://zzz.bwh.harvard.edu/plink/dataman.shtml#recode</a></p> <p><strong>1) In the form of set of files compatible with PLINK out of the box</strong></p> <p>The archive file is called <a href="https://zenodo.org/api/files/b34fd40e-2db1-47b0-92c8-0ad51ad92d46/AtPolyDB_call_method_75_Horton2012.tar.gz">AtPolyDB_call_method_75_Horton2012.tar.gz</a> and contains three files:</p> <ul> <li>genotype.ped: pedigree information from the 1307 ecotypes</li> <li>genotype.map: the SNP positions on the genome</li> <li>phenotypes.pheno: the phenotype value of the 1307 ecotypes</li> </ul> <p> </p>
For MACHINE LEARNING DATABASE evaluation: old-version SEM images of TiO2 particles UNITO
Test images recorded with old ZEISS software, metadata version could differ to the up-to-date version.
Supporting data for "Nuclear quantum effects on zeolite proton hopping kinetics explored with machine learning potentials and path integral molecular dynamics"
<p>Supporting data for "<a href="https://www.nature.com/articles/s41467-023-36666-y">Nuclear quantum effects on zeolite proton hopping kinetics explored with machine learning potentials and path integral molecular dynamics</a>" by M. Bocus, R. Goeminne, A. Lamaire, M. Cools-Ceuppens, T. Verstraelen and V. Van Speybroeck, <em>Nature Communications</em>, <strong>2023</strong>, 14, 1008.</p> <p>This dataset contains examples of input files, submission and analysis scripts to train and use a machine learning potential based on the Schnet architecture for the proton hopping reaction in the H-CHA zeolite. The complete DFT training set, obtained by unbiasing the forces printed by CP2K (with PLUMED coupling), is stored as extended xyz files in the folders DFT/A-B/training_data.xyz where A=1-3 and A<B<5. More details on the folder architecture can be found in the README.md file.</p>
OUTCAR dataset for machine learning potential about a h-BN growth on Pt(111) surface
<p>The growth of monolayer h-BN from boron and nitrogen atoms on Pt(111) is investigated using molecular dynamics combined with machine-learning potentials trained based on first-principles data. The MD simulation can be performed to investigate the h-BN growth on the Pt(111) surface. The training dataset and machine learning potential have been made by the active learning method [1].</p> <p> </p> <p>[1] L. Zhang, D.-Y. Lin, H. Wang, R. Car, E. Weinan, Active learning of uniformly accurate interatomic potentials for materials simulation, Physical Review Materials 3 (2019) 023804.</p>
cropped diBas dataset for machine learning
<p>This zip file contains a collection of 244 x 244 images of various gram-stained bacteria strains, cropped from the diBas dataset (Bartosz Zieli ́nski et al. 2017). Each folder is named by the species of the bacteria images it contains. </p> <p>Potential uses for the dataset include machine learning. An example neural network for bacteria classification using this dataset is deployed at <a href="https://huggingface.co/spaces/qile0317/Bacteria-Classification">https://huggingface.co/spaces/qile0317/Bacteria-Classification</a></p> <p>Note that the data is not completely optimized, as it contains trace amounts of almost blank images. </p>
A machine learning based prediction model for life expectancy
<p>The social and financial systems of many nations throughout the world are significantly impacted by life expectancy (LE) models. Numerous studies have pointed out the crucial effects that life expectancy projections will have on societal issues and the administration of the global healthcare system. The computation of life expectancy has primarily entailed building an ordinary life table. However, the life table is limited by its long duration, the assumption of homogeneity of cohorts and censoring. As a result, a robust and more accurate approach is inevitable. In this study, a supervised machine learning model for estimating life expectancy rates is developed. The model takes into consideration health, socioeconomic, and behavioral characteristics by using the eXtreme Gradient Boosting (XGBoost) algorithm to data from 193 UN member states. The effectiveness of the model's prediction is compared to that of the Random Forest (RF) and Artificial Neural Network (ANN) regressors utilized in earlier research. XGBoost attains an MAE and an RMSE of 1.554 and 2.402, respectively outperforming the RF and ANN models that achieved MAE and RMSE values of 7.938 and 11.304, and 3.86 and 5.002, respectively. The overall results of this study support XGBoost as a reliable and efficient model for estimating life expectancy.</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>
Code and extensive data for training neural networks for radiation, used in "Implementation of a machine-learned gas optics parameterization in the ECMWF Integrated Forecasting System: RRTMGP-NN 2.0""
<p>Data and code used in a paper submitted to JAMES titled :<em> Implementation of a machine-learned gas optics parameterization in the ECMWF Integrated Forecasting System</em></p> <p>1) The files <strong>ml_training_*.7z</strong> contain extensive datasets (in NetCDF format) for training neural network versions of the RRTMGP gas optics scheme as described in the paper. The datasets are read by <a href="https://github.com/peterukk/rte-rrtmgp-nn/blob/main/examples/rrtmgp-nn-training/ml_train.py">ml_train.py.</a></p> <p>2) The ML datasets were in turn generated using the input profiles (in NetCDF format) inside <strong>inputs_to_RRTMGP.zip </strong>by running the Fortran programs <code>rrtmgp_sw_gendata_rfmipstyle.F90 and rrtmgp_lw_gendata_rfmipstyle.F90 </code>in <em>rte-rrtmgp-nn/examples/rrtmgp-nn-training</em>, which call the RRTMGP gas optics scheme, The input profiles contain <strong>millions of columns, hundreds of perturbation experiments (including hypercube-sampled gas concentrations), are derived from several different data sources (including CAMS reanalysis, GCM, and CKDMIP-MMM), and span present-day, preindustrial, and future atmospheric conditions.</strong> They could be used to generate training data for developing emulators of the full RTE+RRTMGP radiation scheme, not just gas optics (see nn_dev on the <a href="https://github.com/peterukk/rte-rrtmgp-nn">RTE+RRTMGP-NN repository on Github</a>, used in a previous paper where different emulation methods were compared)</p> <p>3) The Fortran and Python code used for data generation and NN training are found in<a href="https://github.com/peterukk/rte-rrtmgp-nn/tree/main/examples/rrtmgp-nn-training"> <em>rte-rrtmgp-nn/examples/rrtmgp-nn-training</em> </a>on the main branch on Github; <strong>an archived version is also included here </strong>(<strong>rte-rrtmgp-nn-2.0.zip</strong>). See the readme in the above sub-directory for further information.</p> <p> </p>
Data Extraction table for the study Machine-based Stereotypes: How Machine Learning Algorithms Evaluate Ethnicity from Face Data
<p>This table contains the data extraction results for the study Machine-based Stereotypes: How Machine Learning Algorithms Evaluate Ethnicity from Face Data. It contains 24 columns and 74 rows.</p>
Air quality data from the article "Typhoon-associated air quality over the Guangdong–Hong Kong–Macao Greater Bay Area, China: machine-learning-based prediction and assessment"
<p>This dataset consists of 26 files. The descriptions of the files are as follows:</p> <ul> <li>aqi_TY.csv, pm25_TY.csv, pm10_TY.csv, so2_TY.csv, no2_TY.csv and o3_TY.csv are the observed values of AQI and concentrations of PM<sub>2.5</sub>, PM<sub>10</sub>, SO<sub>2</sub>, NO<sub>2</sub> and O<sub>3</sub> of 36 monitoring stations used in model establish stage on TY days. The time range is June 2014 to December 2020.</li> <li>aqi_NTY.csv, pm25_NTY.csv, pm10_NTY.csv, so2_NTY.csv, no2_NTY.csv and o3_NTY.csv are the observed values of AQI and concentrations of PM<sub>2.5</sub>, PM<sub>10</sub>, SO<sub>2</sub>, NO<sub>2</sub> and O<sub>3</sub> of 36 monitoring stations used in model establish stage on NTY days. The time range is June 2014 to December 2020.</li> <li>station_info.csv is the detailed information of the 36 monitoring stations used in model establish stage, including station number, city, longitude and latitude.</li> <li>aqi_TY_testing.csv, pm25_TY_testing.csv, pm10_TY_testing.csv, so2_TY_testing.csv, no2_TY_testing.csv and o3_TY_testing.csv are the observed values of AQI and concentrations of PM<sub>2.5</sub>, PM<sub>10</sub>, SO<sub>2</sub>, NO<sub>2</sub> and O<sub>3</sub> of 3 monitoring stations used for testing the model on TY days. The time range is June 2014 to December 2020.</li> <li>aqi_NTY_testing.csv, pm25_NTY_testing.csv, pm10_NTY_testing.csv, so2_NTY_testing.csv, no2_NTY_testing.csv and o3_NTY_testing.csv are the observed values of AQI and concentrations of PM<sub>2.5</sub>, PM<sub>10</sub>, SO<sub>2</sub>, NO<sub>2</sub> and O<sub>3</sub> of 3 monitoring stations used for testing the model on NTY days. The time range is June 2014 to December 2020.</li> <li>sta_testing.csv is the detailed information of the 3 monitoring stations used for testing the model, including station number, city, longitude and latitude.</li> </ul>
Machine learning deconvolution of the immune response to dengue
<p>Machine learning deconvolution of the immune response to dengue - dataset of antibody repertoire sequencing</p>
Many-body machine learning models for water, acetonitrile, and methanol
<p><a href="https://keithgroup.github.io/mbGDML/">GDML</a>, <a href="https://libatoms.github.io/GAP/">GAP</a>, and <a href="https://schnetpack.readthedocs.io/en/stable/">SchNet</a> models trained on 1-, 2-, and 3-body energies and forces of water, acetonitrile, and methanol. Size-transferable <a href="https://github.com/mir-group/nequip">NequIPs</a> are trained on trimer data. Energies and forces were computed at the MP2/def2-TZVP level of theory in ORCA v4.2.0. Data sets, training scripts, and analyses of these potentials are available <a href="https://github.com/keithgroup/mbgdml-h2o-meoh-mecn">here</a>. Applications of these models on molecular dynamics simulations are found <a href="https://doi.org/10.5281/zenodo.7112198">here</a>.</p> <p><strong>Changelog</strong></p> <p>The format is based on <a href="https://keepachangelog.com/en/1.0.0/">Keep a Changelog</a>, and this project adheres to <a href="https://semver.org/spec/v2.0.0.html">Semantic Versioning.</a></p> <p>[0.0.2] - 2022-12-20</p> <p>Added</p> <ul> <li><a href="https://github.com/mir-group/nequip">NequIPs</a> trained for all solvents using 1000 trimers.</li> </ul> <p>[0.0.1] - 2022-09-25</p> <ul> <li>Initial release!</li> </ul> <p> </p>
Blood pressure monitoring during anesthesia induction using PPG morphology features and machine learning
<p>PPG-BP dataset of forty patients undergoing general anesthesia, as described in the corresponding journal publication at PLOS ONE (10.1371/journal.pone.0279419).</p> <p>When using this data, please cite the corresponding journal publication.</p> <p> </p>
Identifying serpentine minerals by their chemical compositions with machine learning (dataset and python code)
<p>The dataset and python code for the manuscript of Identifying serpentine minerals by their chemical compositions with machine learning (submitted to American Geologist)</p>
Artifact for "MAAT: A Novel Ensemble Approach to Addressing Fairness and Performance Bugs for Machine Learning Software"
<p>This artifact is for the paper entitled “MAAT: A Novel Ensemble Approach to Addressing Fairness and Performance Bugs for Machine Learning Software”, which is accepted by ESEC/FSE 2022. MAAT is a novel ensemble approach to improving the fairness-performance trade-off for ML software. It outperforms state-of-the-art bias mitigation methods. The artifact has also been placed on GitHub (https://github.com/chenzhenpeng18/FSE22-MAAT) under the Apache License, publicly accessible to other researchers. In this artifact, we provide the source code of MAAT and other existing bias mitigation methods that we use in our study, as well as the intermediate results, the installation instructions, and a replication guideline (included in the README). The replication guideline provides detailed steps to replicate all the results for all the research questions.</p>
Coupled Model Machine Learning Weights
<p>The coupled model weights file contains the necessary weights to the machine learning network to run our machine learning coupled model. The regridded ERA5 file contained a year's worth of initial conditions to be used for starting forecasts or climate simulations. </p>
Curated Dataset of Association Constants Between a Cyclodextrin and a Guest for Machine Learning: Raw Data and Generation Script
<p>Determining the association constant between a cyclodextrin and a guest molecule is an important task for various applications in various industrial and academical fields. However, such a task is time consuming, tedious and requires samples of both molecules. A significant number of association constants and relevant data is available from the literature. The availability of data makes the use of machine learning techniques to predict association constants possible. However, such data is mainly available from tables in articles or appendices. It is necessary to make them available in a computer friendly format and to curate them. Furthermore, the raw data need to be enriched with physicochemical information about each molecule and when such information does not allow to discriminate molecules, some additional data is needed. We present a dataset built from data gathered from the literature. The dataset contains both the original raw data from the articles and the enriched ones. We also provide the scripts used to curate and enrich the raw data.</p>
[Replication package] Explainable Human-Machine Teaming using Model Checking and Interpretable Machine Learning
<p>Anonymized replication package of submission #1917: "Explainable Human-Machine Teaming using Model Checking and Interpretable Machine Learning".</p> <p>See README.md for further instructions.</p>
Data for the paper "Machine-learning-assisted Monte Carlo fails at sampling computationally hard problems"
<p>Data for the paper "Machine-learning-assisted Monte Carlo fails at sampling computationally hard problems"</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.