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.
921
datasets available to search
ShareScore release 0.7.1
Dataset results
921 results for “neural networks”
Dataset for "Comparing Neural Network Based Segmentation of Cardiomyocytes from Different Histology Data"
<p>This dataset contains the trained UNet TensorFlow network, the raw (tiled) histology images, and input and output inference data. The data was used for the work presented at the ISMRM Workshop on Machine Learning in 2018.</p>
Visual perception of liquids: insights from deep neural networks
<p>Datasets and analysis code of the following publication:</p> <p>Van Assen, J.J.R., Nishida, S. & Fleming, R. W. (2020). Visual perception of liquids: insights from deep neural networks. <em>PLOS Computational Biology. </em>DOI: 10.1371/journal.pcbi.1008018</p> <p>For any questions please contact the first author at mail [at] janjaap [dot] info</p> <p><strong>Contents:</strong></p> <p>1. DataAnalysis<br> - Jupyter Notebook to run the full analysis in R<br> - For installation details see: https://irkernel.github.io/requirements/</p> <p>2. FullStimulusSet<br> - 2 million liquid images with 16 viscosities, 10 scenes, 625 variations, and 20 frames<br> - Matlab script that merges the images horizontally for network input</p> <p>3. NeuralActivations<br> - Matlab files containing the neural activations if you cannot read out the networks</p> <p>4. TrainedNetworks<br> - 100 Trained networks referred to in the paper using Matlab and the Deep Learning Toolbox<br> - One custom layer file “switchLayerAdvanced.m”</p> <p>5. ValidationSet<br> - 800 experimental stimuli that were used for validation 16 viscosities, 10 scenes, 5 variations (1,6,11,16,21)<br> - Matlab script that merges the images horizontally for network input</p>
Theater questionnair data and BP neural network prediction network
<p>More than 1500 questionnaires were distributed to the three theaters, and a total of 1382 valid questionnaires were collected for this study: 416 in small-scale theater, 478 in medium-scale theater, 488 in large-scale theater.</p>
Replication data for Convolutional neural networks with hierarchical context transfer for high-resolution spatiotemporal predictions
<p>Data represents number of posts from Instagram in area for six large cities (New York, London, Moscow, Vienna, Tokyo, and Saint Petersburg) covering 2017 and 2018. We aggregated data by hours and split each city using hierarchical area split by 10x10 grid with three levels. The size of the cell on the micro-level equals to 50 meters. Thus, we get 8760 examples for each year. Cells with zero number of posts in an hour are not listed.</p> <p>Data stored in JSON files with the following format:<br> {"2018-12-13 04:00:00":<br> {"(4, 4)":<br> {"sum":1,<br> "data":<br> {"(9, 6)":<br> {"sum":1,<br> "data":{"(2, 4)":1}}}}}}</p>
Cross-Node Federated Graph Neural Network for Spatio-Temporal Data Modeling
<p>Dataset for ICLR 2021 submission "Cross-Node Federated Graph Neural Network for Spatio-Temporal Data Modeling".</p>
Metrics of neural network models in 2022 crops in Extremadura, Spain
<p>Characteristics of the 32 neural network models generated through the analysis of Sentinel-2 images in crops in Extremadura in 2022.</p>
Interpretable Deep Convolutional Neural Network-based Surrogates for Complex Urban Hydrodynamic Modeling using Random Chaotic Rainfall
<p>This data is part of the article "Interpretable Deep Convolutional Neural Network-based Surrogates for Complex Urban Hydrodynamic Modeling using. Random Chaotic Rainfall "data for model training and validation, as well as a dynamic runoff range generated by DHMUrban</p>
[LS2N_IPI_DisFER] Comparing the Robustness of Humans and Deep Neural Networks on Facial Expression Recognition
<h2>DisFER</h2> <p>Distorted-FER (DisFER), a new facial expression recognition (FER) dataset composed of a wide number of distorted images of faces.</p> <p> </p> <h2>Materials and Methods</h2> <h4>Dataset</h4> <div>The source images used in our experiment come from the Facial Expression Recognition 2013 (FER-2013) dataset [1]. This dataset was firstly introduced in 2013 at the International Conference on Machine Learning, and has been used in a large number of research works since then, as it encompasses naturalistic conditions and challenges. This dataset consists of 35,887 images of faces in 48 × 48 format, collected thanks to a Google search. Human accuracy on FER-2013 was estimated by its authors around 65.5% [1].</div> <div>To build the Distorted-FER (DisFER) dataset, we randomly selected, from FER-2013, twelve images per basic emotion, as defined by Ekman [2] (i.e., anger, disgust, fear, happiness, neutral, sadness, and surprise). This yields a total of 84 source images. Each original stimulus was then distorted using three different types of distortions, i.e., Gaussian blur (GB), Gaussian noise (GN), and salt-and-pepper noise (SP). Each distortion was applied at distinct levels: three standard deviation values were tested for GB, i.e., 0.8, 1.1, and 1.4; similarly for GN with standard deviation values equal to 10, 20, and 30; while probability levels of 0.02, 0.04, and 0.06 were chosen for SP; corresponding to low, medium, and high distortions, respectively. </div> <div> </div> <div> <h4>Crowdsourcing Experiment</h4> <div>In order to collect as many votes as possible on our dataset, and because rating 840 images is time-consuming and can be extremely tiring for a single participant, we decided to set up a crowdsourcing experiment. Such experiments indeed allow the conduct of large-scale subjective tests with reduced costs and efforts.</div> <div>The DisFER dataset was therefore split into twenty-one playlists of forty images each, with a view to keep the tests as fast as possible—as crowdsourcing experiments should not last more than ten minutes or so. Playlists were carefully designed to contain the same numbers of images of a given configuration (i.e., emotion, distortion types, and distortion levels). Among a playlist, images were randomly displayed to participants.</div> <div>Each participant was asked to choose which emotion (i.e., anger, disgust, fear, happiness, neutral, sadness, or surprise) they recognized in the displayed image. No time constraint was imposed on participants to fulfill the task.</div> <div> </div> <div>A total of 1051 participants (including 50% of females) were recruited using the Prolific platform [3]. Prolific takes into consideration researchers’ needs by maintaining a subject recruitment process that is similar to that of a laboratory experiment. Indeed, participants are fully informed that they are being recruited for a research study. Consequently, this platform allows researchers to eliminate ethical concerns, and it further improves the reliability of collected data.</div> <div> </div> Participants were aged between 19 and 75 years old (with a mean of 30±8.53 -- note that three participants did not wish to respond). Twenty playlists out of twenty-one were entirely watched and rated by fifty distinct participants, whereas one playlist was watched and evaluated by fifty-one participants.</div> <div> </div> <div> <h2>References</h2> </div> <div>[1] Goodfellow, I.J.; Erhan, D.; Carrier, P.L.; Courville, A. Challenges in Representation Learning: A Report on Three Machine Learning Contests. In Proceedings of the Neural Information Processing, Daegu, South Korea, 3–7 November 2013; Springer: Berlin/Heidelberg, Germany, 2013; pp. 117–124</div> <div>[2] Ekman, P. An argument for basic emotions. Cogn. Emot. <strong>1992</strong>, 6, 169–200</div> <div>[3] https://www.prolific.com/</div> <div> </div>
Dataset related to article: Equivariant graph neural network interatomic potential for Green-Kubo thermal conductivity in phase change materials
<p>This repository contains the dataset to train and test the GeTe Machine Learning Interatomic Potential (MLIP). The computational details are given in the manuscript. </p>
Schema for the supervised training of a neural network
<p>A scheme for the training of a supervised machine learning algorithm such as a neural network.</p>
Altimetry-derived Gravity Gradients using Spectral Method and Their Performance in Bathymetry Inversion using Back-Propagation Neural Network
<p>Data and accompanying MATLAB scripts for the machine learning section of the paper <em> Altimetry-derived Gravity Gradients using Spectral Method and Their Performance in Bathymetry Inversion using Back-Propagation Neural Network</em>. </p>
Neural Network for Determination of the Substrate Activation in Enzymes
<p>1. Multiwfn_lap.sh -- a Linux script to run calculation of the Laplacian of electron density grid in the plain of a nucleophile atom and a carbonyl group. </p> <p>2. lapZero150DPI.py - a Python script for visualization of the 2D Laplacian of the electron density map</p> <p>3. crop_image.py - a Python script that crops the image </p> <p>4. 2500_crop_dataset_MolInf.ipynb - a Python notebook that trains the CNN with the dataset_crop_2500-MolInf </p> <p>5. dataset_crop_2500-MolInf.7z - a dataset to train the CNN</p> <p>6. model_crop_2500-MolInf.h5 - trained CNN, ready for utilization </p> <p>7. validation_datasets.zip -- an archive that includes additional datasets for the neural network validation (complexes of the Mpro with substrates containing Ser, Thr and Pro at P2 and a complex of the NDM-1 and imipenem)</p>
Effective Molecular Dynamics from Neural-Network Based Structure Prediction Models
<p>Molecular dynamics simulation (61.5 us) data of 28 one- and two-domain proteins from Jussupow & Kaila: Effective Molecular Dynamics from Neural-Network Based Structure Prediction Models</p>
Data for "Towards quantum gravity with neural networks: Solving the quantum Hamilton constraint of U(1) BF theory"
<h2>1. Repository Information</h2> <p>This repository contains the data produced during the work discussed in in the paper "<a href="https://iopscience.iop.org/article/10.1088/1361-6382/ad84af" target="_blank" rel="noopener">Towards quantum gravity with neural networks: Solving the quantum Hamilton constraint of U(1) BF theory</a>". Please refer to this paper for more details on how the data was produced.</p> <p> </p> <h2>2. Citing</h2> <p>In addition to citing this repository, please also cite the paper mentioned above if you use the data. The citations is:</p> <p>[1] Hanno Sahlmann and Waleed Sherif 2024 <em>Class. Quantum Grav.</em> <strong>41</strong> 225014</p> <p> </p> <h2>3. File Description</h2> <p>In this repository, you will find 4 general directories (here called parent directories):</p> <ol> <li>Tabulated Data</li> <li>Misc</li> <li>Entanglement Entropy</li> <li>Appendix Data</li> </ol> <p>Each of these directories correposnd to different data produced and discussed in the corresponding parts in the paper mentioned above (e.g. the directory "Tabulated Data" contains the data used in Table 1 and Table 2 in the paper).</p> <p>Each of these parent directories contain within them several sub-directories (child directories) corresponding to different produced data. The raw data can be found in a <code>.json</code> file inside the child directories.</p> <p> </p> <h2>4. Usage</h2> <h3>4.1 Raw Simulation Data</h3> <p>The <code>.json</code> files include the raw data produced during the study. These files can be easily accessed using a python script, as an example, by using:</p> <p><code>import json</code></p> <p><code>filePath = ...</code></p> <p><code>data = json.load(open(filePath))</code></p> <p>where <code>filePath</code> should hold the correct path to the local data once downloaded. Once loaded, the data is handled as a python <code>dict</code>. The dictionary will have a parent key called "Energy", which in itself is yet another dictionary which will always include the keys:</p> <ul> <li>iters</li> <li>Mean</li> <li>Variance</li> <li>Sigma</li> <li>R_hat</li> <li>TauCorr</li> </ul> <p>Hence, to access the "Mean" values, you use <code>data["Energy"]["Mean"]</code>. The data represents the values during a simulation of typically 500 iterations, hence, each of the keys mentioned above will correspond to an array of 500 items. The <code>iters</code> array includes merely the iteration number. The <code>Mean</code> array includes the value of the expectation value of the constraint at the corresponding iteration. The <code>Variance</code>, <code>Sigma</code>, <code>R_hat</code> and <code>TauCorr</code> includes the values of the variance and error in the expectation value at the given iteration as well as the split R-hat diagnostic and the time correlation also in the given iteration. </p> <p> </p> <h3>4.2 Variational State Data</h3> <p>Additionally, some child directories will include a <code>.npy</code> file, which holds the amplitudes of the variational state for the given simulation. These files should be loaded using numpy in python. For example:</p> <p><code>import numpy as np</code></p> <p><code>filePath = ...</code></p> <p><code>varState = np.load(filePath)</code></p> <p>This will load the amplitudes as an array into the <code>varState</code> variable.</p> <p> </p> <h3>4.3 Fluctuation results</h3> <p>In some child directories, there will be a <code>.txt</code> file which includes the output of the calculation of the expectation value of some operators and their quantum fluctuations. These are only results, and not data, as the data can only be computed during the simulation.</p> <p> </p> <h2>5. Contact</h2> <p>Shall you have any unanswered questions regarding the usage of the data, please contact the author:</p> <p>Waleed Sherif</p> <p>email: waleed.sherif@fau.de</p> <p> </p> <h2>6. References</h2> <p>The data provided in this repository was produced using the <a href="https://github.com/netket" target="_blank" rel="noopener">NetKet</a>[1] package</p> <p>[1] <a href="https://doi.org/10.21468/SciPostPhysCodeb.7" target="_blank" rel="noopener">doi: 10.21468/SciPostPhysCodeb.7</a></p> <p> </p>
Dataset from: Structure-based prediction of protein-nucleic acid binding using graph neural networks
<p>Datasets used for training/evaluating the neural network models described in our article. Additional documentation related to how these datasets were constructed can be found in the github repository https://github.com/jaredsagendorf/pnabind/tree/master/datasets</p>
Neural network predicted phase arrival times for stations within and around San Juan Basin
<p>Neural network predicted phase arrival times for stations within and around San Juan Basin. This dataset is not manually reviewed and may contain a large number of false positives.</p>
Audio samples from "A Physical Intelligent Instrument using Recurrent Neural Networks"
<p>These are the audio samples from chapter 3 and 4 of my thesis "A Physical Intelligent Instrument using Recurrent Neural Networks".</p>
Dataset for efficient modelling of ionic and electronic interactions by resistive memory- based reservoir graph neural network
<p>Dataset for training the resistive memory-based reservoir graph neural network.</p> <p>In the atomic force calculation experiment, <span lang="EN-HK"><span>a Li</span><sub>3</sub><span>PO</span><sub>4</sub><span> dataset is derived from the melting and quenching trajectory via AIMD simulations. The training, validation, and testing datasets consist of 40,000, 5,000, and 5,000 samples, respectively. </span></span></p> <p><span lang="EN-HK"><span>In the Hamiltonian calculation, a dataset </span><span lang="EN-HK">of various graphene (72 atoms) configurations are generated by AIMD simulations at room temperature, with Hamiltonian data calculated via the OpenMX code</span><span lang="EN-HK">.</span><span lang="EN-HK"> <span>The training, validation, and testing datasets consist of 270, 90, and 90 samples (including atomic structure and Hamiltonian matrix), respectively.</span></span></span></p> <p>Code: https://github.com/hustmeng/RGNN.git</p> <p>1-Atomic_force_dataset.zip and 2-Hamiltonian_dataset.zip are original data.</p> <p>3-Graph_atomic_force.zip and 4-Graph_training_Hamiltonian.zip are graphs. </p> <p> </p> <p>References:</p> <p> </p> <p>1. C.W. Park, M. Kornbluth, J. Vandermause, C. Wolverton, B. Kozinsky, J.P. Mailoa, Accurate and scalable graph neural network force field and molecular dynamics with direct force architecture, npj Comput. Mater. 7(1) (2021) 73. https://github.com/ken2403/gnnff.git</p> <p>2. H. Li, Z. Wang, N. Zou, M. Ye, R. Xu, X. Gong, W. Duan, Y. Xu, Deep-learning density functional theory Hamiltonian for efficient ab initio electronic-structure calculation, Nat. Comput. Sci. 2(6) (2022) 367-377. https://github.com/mzjb/DeepH-pack.git</p> <p>3. D. Pfau, J.S. Spencer, A.G.D.G. Matthews, W.M.C. Foulkes, Ab initio solution of the many-electron Schrödinger equation with deep neural networks, Phys. Rev. Res. 2(3) (2020) 033429. https://github.com/google-deepmind/ferminet.git</p> <p> </p>
Graph neural networks for predicting metal–ligand coordination of transition metal complexes
<p>Supporting Information data for associated publication "Graph neural networks for predicting metal–ligand coordination of transition metal complexes".</p>
Neural Network Radiation Emulator (KMA/NIMS), December
<p>The dataset is a part of https://doi.org/10.5281/zenodo.5220712 (December)</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.