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.
165
datasets available to search
ShareScore release 0.9.0
Dataset results
165 results for “prompting”
Prompt (p2102)
<b>-- <a href="https://doi.org/10.5281/zenodo.11582199">Documentation</a> --</b><br><br><u>Name</u>: Prompt<br><u>musiXplora-ID</u>: p2102<br><u>musiXplora-URI</u>: <a href="https://musixplora.de/mxp/p2102">https://musixplora.de/mxp/p2102</a><br><u>Gender</u>: m<br><u>First Mentioned</u>: 1678<br><u>Sectors</u>: Instrumentenbau<br><u>Professions (Historical)</u>: Erfinder des Apollon<br><u>Professions (Musical)</u>: Instrumentenbauer, Lautenist<br><u>Other Places of Activity</u>: Paris<br><br><br><br><u>Changelog</u>:<br> - v0.0.1: Initial Upload.<br>
Comparative wordlist prompts for Australian languages
<p>An aligned version of three wordlists, Sutton & Walsh, Curr, and Bates. See also David Nash's item (https://zenodo.org/record/1476467#.W9uXR3ozbUI) of Various Australian wordlist schemes.</p>
Model quality assessment: Results of human ratings, prompts, models as .csv and answers of the LLM
<p>This is the dataset for the paper "<a href="https://doi.org/10.1007/978-3-031-77908-4_7" target="_blank" rel="noopener">Assessing Model Quality Using Large Language Models</a>" and contains the following files of the assessment of the model quality:</p> <ul> <li>Results of the human ratings</li> <li>Prompts</li> <li>Models as .csv</li> <li>Answers of the LLM</li> </ul> <p>The graphical representation of the models is linked under Related works.</p>
A foundation model enpowered by a multi-modal prompt engine for universal seismic geobody interpretation across surveys
<p>A multi-type geobody dataset for training SAG model, including channel, paloekarst, salt body, and so on.</p> <p>A foundation model enpowered by a multi-modal prompt engine for universal seismic geobody interpretation across surveys (<a href="https://arxiv.org/abs/2409.04962">[2409.04962] A foundation model enpowered by a multi-modal prompt engine for universal seismic geobody interpretation across surveys (arxiv.org)</a>)</p> <p> </p>
Can Developers Prompt? A Controlled Experiment for Code Documentation Generation [Replication Package]
<h2>Artifact Summary</h2> <p>This repository contains the replication package for the paper 'Can Developers Prompt? A Controlled Experiment for Code Documentation Generation,' presented at the <em><a href="https://conf.researchr.org/home/icsme-2024" target="_blank" rel="noopener">40th IEEE International Conference on Software Maintenance and Evolution (ICSME'24)</a></em>.</p> <p>The purpose of the package is to facilitate the verification and reproduction of the study results. It provides all data of the controlled experiment, the developed <em>Visual Studio Code (VS Code)</em> extension, as well as the slides of the conference presentations.</p> <h2>Paper Abstract</h2> <p>Large language models (LLMs) bear great potential for automating tedious development tasks such as creating and maintaining code documentation. However, it is unclear to what extent developers can effectively prompt LLMs to create concise and useful documentation. We report on a controlled experiment with 20 professionals and 30 computer science students tasked with code documentation generation for two Python functions. The experimental group freely entered ad-hoc prompts in a ChatGPT-like extension of Visual Studio Code, while the control group executed a predefined few-shot prompt. Our results reveal that professionals and students were unaware of or unable to apply prompt engineering techniques. Especially students perceived the documentation produced from ad-hoc prompts as significantly less readable, less concise, and less helpful than documentation from prepared prompts. Some professionals produced higher quality documentation by just including the keyword Docstring in their ad-hoc prompts. While students desired more support in formulating prompts, professionals appreciated the flexibility of ad-hoc prompting. Participants in both groups rarely assessed the output as perfect. Instead, they understood the tools as support to iteratively refine the documentation. Further research is needed to understand which prompting skills and preferences developers have and which support they need for certain tasks.</p> <h2>References</h2> <p>The published paper is available on <a href="https://doi.org/10.1109/ICSME58944.2024.00058" target="_blank" rel="noopener">IEEE Xplore</a> and the preprint on <a href="https://doi.org/10.48550/arXiv.2408.00686" target="_blank" rel="noopener">arXiv</a>.</p>
EMG from Combination Gestures following Visual Prompts
<p>Dataset of surface EMG recordings from 10 subjects performing single and combination gestures, from Fast and Expressive Gesture Recognition using a Combination-Homomorphic Electromyogram Encoder.</p> <p>Each subject contributes 1224 gesture examples (584 single gestures and 640 combination gestures).</p> <p>For more details and example usage, see the following:</p> <ul> <li>Paper pdf - <a href="https://arxiv.org/pdf/2311.14675.pdf">https://arxiv.org/pdf/2311.14675.pdf</a></li> <li>Experiment code - <a href="https://github.com/nik-sm/com-hom-emg">https://github.com/nik-sm/com-hom-emg</a></li> </ul> <h2>Contents</h2> <p>Dataset of single and combination gestures from 10 subjects. Each subject contributes 1224 gesture examples (584 single gestures and 640 combination gestures).</p> <h2>Data</h2> <p>Each gesture trial consists of 500 ms of activity from 8 recording electrodes around the mid forearm.</p> <p>Recording was performed at 1926 Hz with built-in 20–450 Hz bandpass filtering applied.</p> <h3>Data Collection Details</h3> <p>In each gesture trial, subjects began from a neutral position, with the arm on an armrest. Subjects followed visual cues to perform either a direction gesture, a modifier gesture, or both. The timing structure and number of gestures obtained from each visual prompt varied across experimental blocks.</p> <p>Data for a single gesture trial come from the central 500ms when that gesture is being performed (the window is centered to be as far from gesture onset and gesture end as possible).</p> <p>See the paper for more details.</p> <h2>Labels</h2> <p>Each gesture trial in the dataset has a two-part label, describing which gesture was performed.</p> <p>The first label component describes the direction gesture, and takes values in {0, 1, 2, 3, 4}, with the following meaning:</p> <ul> <li>0 - "Up" (wrist extension)</li> <li>1 - "Down" (wrist flexion)</li> <li>2 - "Left" (wrist abduction; movement towards thumb side)</li> <li>3 - "Right" (wrist adduction; movement towards pinky side)</li> <li>4 - "NoDirection" (absence of a direction gesture; none of the above)</li> </ul> <p>The second label component describes the modifier gesture, and takes values in {0, 1, 2, 3, 4}, with the following meaning:</p> <ul> <li>0 - "Pinch" (connecting thumb and index finger)</li> <li>1 - "Thumb" (touching thumb to first knuckle of index finger)</li> <li>2 - "Fist" (all fingers curled)</li> <li>3 - "Open" (all fingers extended)</li> <li>4 - "NoModifier" (absence of a modifier gesture; none of the above)</li> </ul> <p>Each subject provided 1224 total examples, consisting of:</p> <ul> <li>- 584 single gesture examples. (584 = 73 examples, for each of the 4+4 possible single gesture classes)</li> <li>- 640 combination gesture examples. (640 = 40 examples, for each of the 4x4 possible combination gesture classes)<br> </li> </ul> <h3>Examples of Label Structure</h3> <p>Single gestures have labels like (0, 4) indicating ("Up", "NoModifier") or (4, 3) indicating ("NoDirection", "Open").</p> <p>Combination gesture have labels like (0, 0) indicating ("Up", "Pinch") or (2, 3) indicating ("Left", "Open").</p> <h2>Loading data</h2> <p>After downloading and unzipping, the examples below can be used to load data in Python or MATLAB.</p> <p>Data have shape `(items, channels, timesteps)`.</p> <p>Labels have shape `(items, 2)`, where the first coordinate is the direction and the second coordinate is the modifier.<br> </p> <h3>Python example</h3> <blockquote> <p><em># Load data</em></p> <p>import numpy as np</p> <p>from pathlib import Path</p> <p> </p> <p>folder = Path("path/to/combination-gesture-dataset/python")<br> </p> <p>all_data, all_labels = [], []</p> <p>for i in range(10):</p> <p> data = np.load(f"{folder}/subj{i}/data.npy")</p> <p> labels = np.load(f"{folder}/subj{i}/labels.npy")</p> <p> all_data.append(data)</p> <p> all_labels.append(labels)</p> <p>all_data = np.concatenate(all_data, axis=0)</p> <p>all_labels = np.concatenate(all_labels, axis=0)</p> <p> </p> <p>print(all_data.shape) <em># (12240, 8, 962)</em></p> <p>print(all_labels.shape) <em># (12240, 2)</em></p> <p> </p> <p><em># Convert numeric labels to text</em></p> <p>def convert_label(label):</p> <p> directions = ["Up", "Down", "Left", "Right", "NoDirection"]</p> <p> modifiers = ["Pinch", "Thumb", "Fist", "Open", "NoModifier"]</p> <p> return directions[label[0]], modifiers[label[1]]</p> <p> </p> <p>print(convert_label(all_labels[0])) <em># ('NoDirection', 'Fist')</em></p> </blockquote> <p> </p> <h3>MATLAB example</h3> <blockquote> <p>% Load data</p> <p>folder = "path/to/combination-gesture-dataset/matlab";<br> </p> <p>all_data = [];</p> <p>all_labels = [];</p> <p>for i = 0:9</p> <p> data = load(sprintf("%s/subj%d/data.mat", folder, i));</p> <p> labels = load(sprintf("%s/subj%d/labels.mat", folder, i));</p> <p> all_data = [all_data; data.contents];</p> <p> all_labels = [all_labels; labels.contents];</p> <p>end</p> <p> </p> <p>% (Optional) convert labels from 0-indexed to 1-indexed, for easier conversion to text</p> <p>all_labels = all_labels + 1;</p> <p> </p> <p>size(all_data) % (12240, 8, 962)</p> <p>size(all_labels) % (12240, 2)<br> </p> <p>% Convert numeric labels to text</p> <p>directions = ["Up", "Down", "Left", "Right", "NoDirection"];</p> <p>modifiers = ["Pinch", "Thumb", "Fist", "Open", "NoModifier"];</p> <p>convert_label = @(label) [directions(label(1) + 1), modifiers(label(2) + 1)];</p> <p> </p> <p>convert_label(all_labels(1, :)) % "NoDirection" "Fist"</p> </blockquote>
Unlocking Efficiency: Assessing the Impact of Prompt Pay on Government Operations and Financial Transactions in Thailand
<p> </p> <p>1. Promtpay users</p> <p>2. Monetary Aggregates and Components</p> <p> </p> <p>Data collected and processed as part of the ODDEA (Overcoming Digital Divide Between Europe and Southeast Asia) EU research project (<em>Project ID: HORIZON MSCA-SE 101086381)</em></p>
Exploring Changes in COVID-19 Vaccination Intentions by Prompting Altruistic Motives Using a Video Intervention
ClinicalTrials.gov study NCT04960228. IPD Sharing: NO. Countries: 1. Publications: 3.
Evaluation of large language model chatbot responses to psychotic prompts: numerical ratings of prompt-response pairs
Open the record for dataset details and reuse information.
Data from: Seasonal food scarcity prompts long-distance foraging by a wild social bee
Foraging is an essential process for mobile animals and its optimization serves as a foundational theory in ecology and evolution; however, drivers of foraging are rarely investigated across landscapes and seasons. Using a common bumble bee species from the Western US (Bombus vosnesenskii), we ask if seasonal decreases in food resources prompt changes in foraging behavior and space use. We employ a unique integration of population genetic tools and spatially-explicit foraging models to estimate foraging distances and rates of patch visitation for wild bumble bee colonies across three study regions and two seasons. By mapping the locations of 669 wild-caught individual foragers, we find substantial variation in colony-level foraging distances, often exhibiting a 60-fold difference within a study region. Our analysis of visitation rates indicates that foragers display a preference for high-cover destination patches and forage significantly further for these patches, but only in the summer, when landscape-level resources are low. Overall, these results indicate that an increasing proportion of long-distance foraging bouts take place in the summer. As pollinators, the foraging dynamics of wild bees are of urgent concern given the potential impacts of global change on their movement and services. The behavioral shift towards long-distance foraging with seasonal declines in food resources suggests a novel phenologically-directed approach to landscape-level pollinator conservation and increased evaluation of late-season floral resources.
Prompted Textures Dataset (PTD)
<p>The Prompted Textures Dataset (PTD) is a synthetic texture image dataset consisting of 246,285 images across 56 different texture classes from the work <a title="On Synthetic Texture Datasets: Challenges, Creation, and Curation" href="https://www.arxiv.org/abs/2409.10297" target="_blank" rel="noopener">On Synthetic Texture Datasets: Challenges, Creation, and Curation</a>.</p> <p>If you find this dataset useful in your work, please consider citing:</p> <p>@misc{hoak2024synthetictexturedatasetschallenges,<br> title={On Synthetic Texture Datasets: Challenges, Creation, and Curation}, <br> author={Blaine Hoak and Patrick McDaniel},<br> year={2024},<br> eprint={2409.10297},<br> archivePrefix={arXiv},<br> primaryClass={cs.CV},<br> url={https://arxiv.org/abs/2409.10297}, <br>}</p> <p> </p>
Prompts IA generativa: Tendencias, categorías y estadísticas de uso (UOC - PRA1)
<p>El presente dataset ha sido elaborado en el contexto de un proyecto de web scraping de la asignatura de "Tipología y ciclo de vida de los datos" de la Universidad Oberta de Catalunya (UOC).</p><p>El dataset recoge una muestra de los datos capturados del sitio web 'snackprompt.com'. Concretamente, se recoge información relativa a los prompts publicados por los diferentes usuarios que hacen uso de la plataforma.</p>
Data from: Phylogenomic analyses of echinoid diversification prompt a re-evaluation of their fossil record
<p>Echinoids are key components of modern marine ecosystems. Despite a remarkable fossil record, the emergence of their crown group is documented by few specimens of unclear affinities, rendering their early history uncertain. The origin of sand dollars, one of its most distinctive clades, is also unclear due to an unstable phylogenetic context. We employ eighteen novel genomes and transcriptomes to build a phylogenomic dataset with a near-complete sampling of major lineages. With it, we revise the phylogeny and divergence times of echinoids, and place their history within the broader context of echinoderm evolution. We also introduce the concept of a chronospace—a multidimensional representation of node ages—and use it to explore methodological decisions involved in time calibrating phylogenies. We find the choice of clock model to have the strongest impact on divergence times, while the use of site-heterogeneous models and alternative node prior distributions showing minimal effects. The choice of loci has an intermediate impact, affecting mostly deep Paleozoic nodes, for which clock-like genes recover dates more congruent with fossil evidence. Our results reveal that crown group echinoids originated in the Permian and diversified rapidly in the Triassic, despite the relative lack of fossil evidence for this early diversification. We also clarify the relationships between sand dollars and their close relatives and confidently date their origins to the Cretaceous, implying ghost ranges spanning approximately 50 million years, a remarkable discrepancy with their rich fossil record.</p>
Prompts generated from ChatGPT3.5, ChatGPT4, LLama3-8B, and Mistral-7B with NYT and HC3 topics in different roles and parameters configurations
<h2>Description</h2> <p>Prompts generated from ChatGPT3.5, ChatGPT4, Llama3-8B, and Mistral-7B with <a href="https://www.nytimes.com/2023/07/19/learning/175-writing-prompts-to-spark-discussion-and-reflection.html">NYT</a> and <a href="https://huggingface.co/datasets/Hello-SimpleAI/HC3">HC3</a> topics in different roles and parameter configurations. </p> <p>The dataset is useful to study lexical aspects of LLMs with different parameters/roles configurations.</p> <ul> <li>The 0_Base_Topics.xlsx file lists the topics used for the dataset generation</li> <li>The rest of the files collect the answers of ChatGPT to these topics with different configurations of parameters/context: <ul> <li>Temperature (parameter): Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.</li> <li>Frequency penalty (parameter): Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.</li> <li>Top probability (parameter): An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass.</li> <li>Presence penalty (parameter): Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.</li> <li>Roles (context) <ul> <li>Default: No role is assigned to the LLM, the default role is used.</li> <li>Child: The LLM is requested to answer as a five-year-old child. </li> <li>Young adult male: The LLM is requested to answer as a young male adult. </li> <li>Young adult female: The LLM is requested to answer as a young female adult. </li> <li>Elderly adult male: The LLM is requested to answer as an elderly male adult. </li> <li>Elderly adult female: The LLM is requested to answer as an elderly female adult.</li> <li>Affluent adult male: The LLM is requested to answer as an affluent male adult. </li> <li>Affluent adult female: The LLM is requested to answer as an affluent female adult. </li> <li>Lower-class adult male: The LLM is requested to answer as a lower-class male adult. </li> <li>Lower-class adult female: The LLM is requested to answer as a lower-class female adult. </li> <li>Erudite: The LLM is requested to answer as an erudite who uses a rich vocabulary.</li> </ul> </li> </ul> </li> </ul> <h2>Paper</h2> <ul> <li>Paper: <a href="https://dl.acm.org/doi/10.1145/3696459">Beware of Words: Evaluating the Lexical Diversity of Conversational LLMs using ChatGPT as Case Study</a></li> <li>Cite:</li> </ul> <p><code>@article{10.1145/3696459,</code><br><code>author = {Mart\'{\i}nez, Gonzalo and Hern\'{a}ndez, Jos\'{e} Alberto and Conde, Javier and Reviriego, Pedro and Merino-G\'{o}mez, Elena},</code><br><code>title = {Beware of Words: Evaluating the Lexical Diversity of Conversational LLMs using ChatGPT as Case Study</code><code>},</code><br><code>year = {2024},</code><br><code>publisher = {Association for Computing Machinery},</code><br><code>address = {New York, NY, USA},</code><br><code>issn = {2157-6904},</code><br><code>url = {https://doi.org/10.1145/3696459},</code><br><code>doi = {10.1145/3696459},</code><br><code>abstract = ,</code><br><code>note = {Just Accepted},</code><br><code>journal = {ACM Trans. Intell. Syst. Technol.},</code><br><code>month = sep,</code><br><code>keywords = {LLM, Lexical diversity, ChatGPT, Evaluation}</code><br><code>}</code></p>
Molecular dynamics trajectories and portable binary run files for the self-assembly of heparin and amyloid-β(16-22) with the ProMPT forcefield
<h1>About this repository</h1> <p>Self-assembly simulations of heparin and amyloid-β(16-22) were performed at various numbers of peptides (<em>N_pep</em>), number of heparin molecules (<em>N_hep</em>), degrees of polymerization of heparin (<em>hep_dp</em>) and rigidity factors (<em>rig_f</em>) for heparin. This repository contains one folder per system, characterized by a combination of four system variables: <em>N_pep</em>, <em>N_hep</em>, <em>hep_dp</em> and <em>rig_f. </em></p> <p>The simulations were performed on the GROMACS 2019.4 molecular dynamics engine, with the ProMPT forcefield for coarse-grained molecular dynamics. Four independent trials: <em>trial_A</em>, <em>trial_B</em>, <em>trial_C</em> and <em>trial_D</em> were performed per system, each with different starting velocities. Each trial was run for 3000 ns.</p> <h1>Contents</h1> <h2><code>> heparin_abeta_trajectories.zip</code></h2> <p>The repository contains the zip file <code>heparin_abeta_trajectories.zip</code> with 13 folders named according to the convention,<code> "{<em>N_pep</em>}pep_{<em>N_hep</em>}hep_dp{<em>hep_dp</em>}_{<em>rig_f</em>}xRigid"</code>. For example, data for the system consisting of 16 peptides (<em>N_pep</em>), 1 heparin (<em>N_hep</em>), 18 monosaccharides in length (<em>hep_dp</em>) with a rigidity factor of 100 (<em>rig_f</em>) would be stored in the directory <code>16pep_1hep_dp18_100xRigid/</code>.</p> <p>If the system did not contain heparin, <em>N_hep, hep_dp</em> and <em>rig_f</em> were set to 0 by default. For example, data for the system consisting of 16 peptides (<em>N_pep</em>) and no heparin would be stored in the directory <code>16pep_0hep_dp0_0xRigid/</code>.</p> <p>A directory such as <code>16pep_1hep_dp18_100xRigid/</code> will have the following contents:</p> <ul> <li><code><strong>solute_only.ndx</strong></code><br>Contains an index group for solute molecules (peptides and/or heparin) only.</li> <li><code><strong>trial_A/</strong></code> <ul> <li><code><strong>md.tpr</strong></code> <br>Portable run file with which the current trajectory was generated. This file may be used to reproduce the trajectory as well.</li> <li><code><strong>solute_only.cluster_center.xtc </strong></code><br>A gromacs trajectory containing only the solute molecules (peptides and/or heparin), centered with the gromacs tool <em>gmx trjconv</em></li> <li><code><strong>solute_only.tpr </strong></code><br>A gromacs portable binary input file containing data for the solute molecules (peptides and/or heparin) only, generated with the <em>gmx convert-tpr</em> tool.<br>This file may be used during analysis of the solute_only.cluster_center.xtc trajectory.</li> </ul> </li> <li><code><strong>trial_B/</strong></code><br>contents same as <code><strong>trial_A/</strong></code></li> <li><strong><code>trial_C/</code><br></strong>contents same as <code><strong>trial_A/</strong></code></li> <li><strong><code>trial_D/</code><br></strong>contents same as <code><strong>trial_A/</strong></code></li> </ul> <h2><code>> psf_files_for_VMD_visualization.zip</code></h2> <p>This zip file contains thirteen .psf files, one per system, that can be used in accord with <code>solute_only.cluster_center.xtc </code>files to visualize trajectories on the Visual Molecular Dynamics (VMD) software.</p> <p><strong>Please seek out the associated publication for essential context on these trajectories. </strong></p> <p><strong>To access the source files with which these simulations were set-up, and a brief tutorial, see: </strong><a href="https://github.com/suhasgotla/heparin_amyloid_self-assembly">https://github.com/suhasgotla/heparin_amyloid_self-assembly</a></p>
(supplementary material) Fine-Tuning and Prompt Engineering for Large Language Models-based Code Review Automation
<div> <div> <div> <div>Supplementary material for paper <strong>"Fine-Tuning and Prompt Engineering for Large Language Models-based Code Review Automation"</strong></div> <div> </div> <div> <div> <div>The script for the paper can be found in this GitHub repository: https://github.com/awsm-research/LLM-for-code-review-automatiton</div> </div> </div> </div> </div> </div>
Landscape scenicness prompts
<p>Reproduction repository for the paper titled "Prompt-guided and multimodal landscape scenicness assessments with vision-language models". This repo contains the following files:</p> <ol> <li>Prompts collected by our crowdsourcing effort</li> <li>CLIP-and SigLIP extracted embeddings of ScenicOrNot images</li> <li>Splits used for few-shot learning</li> <li>GeoJSON file containing geotagged ScenicOrNot labels for each image</li> </ol> <p>The original ScenicOrNot images are hosted by Geograph, whose license terms do not allow re-sharing of the images. Instead, we make the embeddings used in our study available and provide scripts to download the original images on our <a href="https://github.com/ahlevering/scenicness_prompting">Github repository</a>, which also contains the code used for this study. Alternatively, please contact the authors for a copy of the SON images.</p>
dataset of prompts for development of the optimization guiding framework if-act
Open the record for dataset details and reuse information.
Data package for paper "Transformer models for astrophysical time series and the GRB prompt-afterglow relation"
<p>This is a data package accompanying the paper "Transformer models for astrophysical time series and the GRB<br>prompt-afterglow relation". The code used to acquire the data is in the "data" folder. The code used to analyse the data is in the "analysis" folder.</p> <p>DOI paper: <a href="https://doi.org/10.1093/rasti/rzae026">10.1093/rasti/rzae026</a></p>
Training datasets for "Multi-purpose controllable protein generation via prompted language models"
<div> <p>The datasets used to tune modular prompts of PROPEND fall into three main categories based on their design objectives: tertiary structure, secondary structure, and functional annotation.</p> </div>
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.