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.
598
datasets available to search
ShareScore release 0.9.0
Dataset results
598 results for “classifier”
Liver Micrometastases area quantification using QuPath and pixel classifier
<p><strong>Sample</strong>: Mouse (NSG) liver slices with human colorectal cancer cells metastases, stained with Hematoxylin & Eosin. </p> <p><strong>Image Acquisition</strong>: Images were acquired on an Olympus VS120 Whole Slide Scanner, using a 20x objective (UPLSAPO, N.A. 0.75) and a color camera (Pike F505 Color) with an image pixel size of 0.345 microns.</p> <p><strong>Image Processing and Analysis</strong>: Obtained images were analyzed using the software QuPath [1] (version 0.3.2) using groovy scripts, making use of a pixel classifier to segment and measure cancer cell clusters.</p> <p><strong>Files</strong> :</p> <p><em>Detailed_worflow.pdf</em> : contains a detailed description of how pixel classifier was created</p> <p><em>images_for_classifier_training.zip</em> : contains all the vsi file obtained from the microscope and used for the training</p> <p><em>project_for_classifier_training.zip</em> : contains the QuPath project, with Training Image, annotations, classifiers and scripts for analysis</p> <p><em>PythonCode.txt</em> : code ran to transform output results from QuPath to final results</p> <p> </p> <p>[1] Bankhead, P. et al. <strong>QuPath: Open source software for digital pathology image analysis</strong>. <em>Scientific Reports</em> (2017). <a href="https://doi.org/10.1038/s41598-017-17204-5">https://doi.org/10.1038/s41598-017-17204-5</a></p>
Machine learning classifiers for species classification of fungi using error-prone long-reads on extended metabarcodes
<p>Machine learning models used in the decision tree of linked machine learning models (<a href="https://github.com/teenjes/fungal_ML">https://github.com/teenjes/fungal_ML</a>)</p>
DECIMER Image classifier dataset
<p>Images dataset divided into train (10905114 images), validation (2115528 images) and test (544946 images) folders containing a balanced number of images for two classes (chemical structures and non-chemical structures).</p> <p>The chemical structures were generated using RanDepict to random picked compounds from the ChEMBL30 database and the COCONUT database.</p> <p>The non-chemical structures were generated using Python or they were retrieved from several public datasets:</p> <p>COCO dataset, MIT Places-205 dataset, Visual Genome dataset, Google Open labeled Images, MMU-OCR-21 (kaggle), HandWritten_Character (kaggle), CoronaHack -Chest X-Ray-dataset (kaggle), PANDAS Augmented Images (kaggle), Bacterial_Colony (kaggle), Ceylon Epigraphy Periods (kaggle), Chinese Calligraphy Styles by Calligraphers (kaggle), Graphs Dataset (kaggle), Function_Graphs Polynomial (kaggle), sketches (kaggle), Person Face Sketches (kaggle), Art Pictograms (kaggle), Russian handwritten letters (kaggle), Handwritten Russian Letters (kaggle), Covid-19 Misinformation Tweets Labeled Dataset (kaggle) and grapheme-imgs-224x224 (kaggle).</p> <p>This data was used to build a CNN classification model using as a base model EfficienNetB0 and fine tuning it. The model is available on <a href="https://github.com/Iagea/CNN_chem_not_chem">Github</a>.</p>
Lunar Reconnaissance Orbiter Imagery for LROCNet Moon Classifier
<p><strong>Summary</strong></p> <p>We provide imagery used to train LROCNet -- our Convolutional Neural Network classifier of orbital imagery of the moon. Images are divided into train, validation, and test zip files, which contain class specific sub-folders. We have three classes: "fresh crater", "old crater", and "none". Classes are described in detail in the attached labeling guide.</p> <p><strong>Directory Contents</strong></p> <p>We include the labeling guide and training, testing, and validation data. Training data was split to avoid upload timeouts.</p> <ul> <li>LROC_Labeling_Intro_for_release.ppt: Labeling guide</li> <li>val: Validation images divided into class sub-folders <ul> <li>ejecta: "fresh crater" class</li> <li>oldcrater: "old crater" class</li> <li>none: "none" class</li> </ul> </li> <li>test: Testing images divided into class sub-folders <ul> <li>ejecta: "fresh crater" class</li> <li>oldcrater: "old crater" class</li> <li>none: "none" class</li> </ul> </li> <li>ejecta_train: Training images of "fresh crater" class</li> <li>oldcrater_train: Training images of "old crater" class</li> <li>none_train1-4: Training images of "none" class (divided into 4 just for uploading)</li> </ul> <p><strong>Data Description </strong></p> <p>We use CDR (Calibrated Data Record) browse imagery (50% resolution) from the Lunar Reconnaissance Orbiter's Narrow Angle Cameras (NACs). Data we get from the NACs are 5-km swaths, at nominal orbit, so we perform a saliency detection step to find surface features of interest. A detector developed for Mars HiRISE (Wagstaff et al.) worked well for our purposes, after updating based on LROC NAC image resolution. We use this detector to create a set of image chipouts (small 227x277 cutouts) from the larger image, sampling the lunar globe.</p> <p><strong>Class Labeling</strong></p> <p>We select classes of interest based on what is visible at the NAC resolution, consulting with scientists and performing a literature review. Initially, we have 7 classes: "fresh crater", "old crater", "overlapping craters", "irregular mare patches", "rockfalls and landfalls", "of scientific interest", and "none".</p> <p>Using the Zooniverse platform, we set up a labeling tool and labeled 5,000 images. We found that "fresh crater" make up 11% of the data, "old crater" 18%, with the vast majority "none". Due to limited examples of the other classes, we reduce our initial class set to: "fresh crater" (with impact ejecta), "old crater", and "none".</p> <p>We divide the images into train/validation/test sets making sure no image swaths span multiple sets.</p> <p><strong>Data Augmentation</strong></p> <p>Using PyTorch, we apply the following augmentation on the training set only: horizontal flip, vertical flip, rotation by 90/180/270 degrees, and brightness adjustment (0.5, 2). In addition, we use weighted sampling so that each class is weighted equally. The training set included here does not include augmentation since that was performed within PyTorch.</p> <p><strong>Acknowledgements</strong></p> <p>The author would like to thank the volunteers who provided annotations for this data set, as well as others who contributed to this work (as in the Contributor list). We would also like to thank the PDS Imaging Node for support of this work.</p> <p>The research was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration (80NM0018D0004).</p> <p>CL#22-4763</p> <p>© 2022 California Institute of Technology. Government sponsorship acknowledged.</p>
Magnetosheath Jets MMS1 (5/2015 - 6/2020) - Classified
<p><strong>README</strong></p> <p>This dataset contains the time and the class of magnetosheath jets observed by MMS 1 during 05/2015 – 06/2020. </p> <p>More information about the different classes can be found in the articles:</p> <ol> <li>https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2019JA027754</li> <li>https://www.frontiersin.org/articles/10.3389/fspas.2020.00024/full</li> </ol> <p>While an extension of the classification process has been published in:</p> <ul> <li>https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2021JA029269</li> </ul> <p><strong>INFO</strong> </p> <p>For the attached text file. The first column is the time of maximum dynamic pressure in UTC. The second column is the class of the jet:</p> <p><em>Main Categories:</em><br>11 = Quasi-parallel jet<br>22 = Quasi-perpendicular jet <br>3 = Boundary jet<br>5 = Encapsulated jet </p> <p><em>Secondary Categories:</em><br>1 = Possibly Quasi-parallel jet<br>2 = Possibly Quasi-parallel jet<br>4 = Possibly Boundary jet<br>6 = Possibly Encapsulated jet<br>7 = Close to Magnetopause or Bow Shock jet<br>0 = Unclassified jet<br>8 = Data gap jet</p> <p>Burst availability is given in the third column as:</p> <p>2 = Full burst availability<br>1 = Partial burst availability<br>0 = No burst availability</p> <p>For more information, please contact the author (savvasraptis@gmail.com) </p>
D3 annotation with CSO Classifier
<p>The <a href="https://zenodo.org/record/7069915">DBLP Discovery Dataset </a>(D3) is a newly created dataset of research papers in the field of Computer Science which can support several tasks like identifying trends in research activity, productivity, focus, bias, accessibility, and impact. This dataset stems from DBLP and integrates additional information from the full-texts. We argue that papers classified with their research topics can improve the identification of research trends. To this end, we used the <a href="https://github.com/angelosalatino/cso-classifier">CSO Classifier</a> to annotate all the papers within D3 and we made such extension available for research purposes.</p> <p> </p> <p>More info: <a href="https://www.salatino.org/wp/annotating-d3-dataset-with-the-cso-classifier/">https://www.salatino.org/wp/annotating-d3-dataset-with-the-cso-classifier/</a></p> <p>More info pdf: <a href="https://www.salatino.org/wp/wp-content/uploads/2022/09/Annotating-D3-dataset-with-the-CSO-Classifier.pdf">https://www.salatino.org/wp/wp-content/uploads/2022/09/Annotating-D3-dataset-with-the-CSO-Classifier.pdf</a></p>
Data from: A semi-automated approach to classify and map ecological zones across the dune-beach interface
<p>This is the raw data behind the publication: </p> <p><strong>A semi-automated approach to classify and map ecological zones across the dune-beach interface</strong></p> <p><strong>Abstract: </strong>Habitat classification and mapping underpins most conservation and management tools, because habitats are often used as a surrogate for all biodiversity. Some habitat boundaries are easy to delineate; however, sandy shores are ecotones or ecoclines given their dynamic interface between the marine and the terrestrial realms. Although methods for mapping habitats along shorelines have been broadly applied, we aim to test a semi-automated approach to mapping across-shore “sub-environments” in this transition zone at a finer scale. Using an empirical dataset of photographs covering a small area (three across-shore transects from each of two different areas) with a high resolution, we tested seven machine learning algorithms to determine which one had the best classification accuracy, and to identify which environmental variables are the main determinants of classifications. The randomForest, stochastic gradient boosting, and C5.0 algorithms most accurately classified the photographs as the correct sub-environment. Based on the randomForest algorithm, the variables entropy, drift cover rate, local slope, segmented vegetation cover and number of points with sand or marine litter had the highest influence on the classification. There was no sensitivity to spatial variation alongshore. This approach can be used to map sub-environments at larger scales using drone technology to capture georeferenced digital photographs systematically. Consequently, coastal habitats can be mapped at a finer scale without causing disturbance to this especially sensitive ecotone.</p>
Improving the drug discovery process by using multiple classifier systems
<p>High-quality dataset gathered from ChEMBL version 22 based on UniProt accession P34972. Regarding to activity data potential, duplicates were ignored, no activity or data validity comments were allowed, only data from binding assays and with a pCheMBL value were kept. This led to a dataset composed of 3925 chemical compounds (instances) represented using 2132 features. The first 2048 features epitomize different chemical structures fingerprints (represented using FCFP_6 notation), while the remaining 84 are associated with several physicochemical descriptors (such as Fractional Polar Surface Area, Rotatable Bonds or Molecular Weight). Finally, the set was transformed into a binary classification set where the activity cut-off was defined at a pChEMBL value > 7 and written to a tab-delimited text file. The final set contained 1977 active compounds and 1948 inactive compounds. Table 3 shows the codification of each feature grouped by type.</p>
Hammer-classified LOTUS dataset
<p>Hammer-classified <a href="https://doi.org/10.5281/zenodo.7534071">LOTUS dataset v10</a>.</p> <p>Hammer is available at <a href="https://github.com/LucaCappelletti94/hammer">https://github.com/LucaCappelletti94/hammer</a>. The current dataset was classified using the `npc.harmonized.v1` version and codebase at the following commit <a href="https://github.com/LucaCappelletti94/hammer/commit/8f808604c18625e4cee4b75d3ca255af259d76a3">8f808604c18625e4cee4b75d3ca255af259d76a3</a>.</p> <p>Classifications available as .csv at the class, superclass and pathway level.</p> <p>Command runned :</p> <div> <div>hammer predict \</div> <div>--input "./downloads/taxo_db_metadata.csv" \</div> <div>--version npc.harmonized.v1 \</div> <div>--verbose \</div> <div>--output-dir "./downloads/npc.harmonized.v1/lotus"</div> </div> <p> </p> <p>Change Log</p> <p>------------</p> <p>v0.2</p> <ul> <li>individual .gz</li> </ul> <p> </p>
Classifying the generation and formation channels of dynamically-formed gravitational-wave events
<p>This dataset contains all the simulations of dynamically-formed binaries performed with the software <a href="https://github.com/Kkritos/Rapster">rapster</a>, together with trained machine-learning classification models from (Antonelli, Kritos et al, in prep.), see <a href="https://github.com/aantonelli94/TheBHClassifier">the public codes online</a>.</p> <p>All items starting with "mergers_*" are simulations of clusters and they follow the structure reported in the documentation of <a href="https://github.com/Kkritos/Rapster">rapster</a>. The simulations differ in the choice of the hyperparameters for the distribution of the cluster mass, half-mass radius and initial spin distribution for the binaries.</p> <p>All items starting from "RFClassifier_*" are machine-learning classification models that use a Random Forest Classifier and that are trained with the simulations above. The models ending with "*_gen" predict the generation of the black holes, those with "*_form" predict their formation channels. </p> <p> </p>
Data on grammatical gender and numeral classifiers
<p>This material contains the dataset from the <a href="https://version.helsinki.fi/hals/sinnemaki/sinnemaki2019">gitlab repository</a> of the following article. Please cite the article when using the data.</p> <p>Sinnemäki, Kaius. 2019. On the distribution and complexity of gender and numeral classifiers. In Di Garbo, Francesca, Bernhard Wälchli & Bruno Olsson (eds.), <em>Grammatical Gender and Linguistic Complexity, Volume II: World-wide Comparative Studies</em> (Studies in Diversity Linguistics 27), 133–200. Berlin: Language Science Press.</p>
The patccat classifier for patent claims - EPO edition
<p>!!! This is the <strong>EPO/European version</strong> of the patccat classifier of patent claims. !!!</p> <p>Note: We use the same approach that we use for USPTO patents. For a detailed description, see <a href="https://doi.org/10.5281/zenodo.6395307">https://doi.org/10.5281/zenodo.6395307</a>.</p> <p><strong>Data version: 3.4.0</strong></p> <p>Authors:<br> Bernhard Ganglmair (University of Mannheim, Department of Economics, and ZEW Mannheim)<br> W. Keith Robinson (Wake Forest University, School of Law)<br> Michael Seeligson (Southern Methodist University, Cox School of Business)</p> <p>Please cite the following paper when using the data in your own work:</p> <p>Ganglmair, Bernhard, W. Keith Robinson, and Michael Seeligson (2022): "The Rise of Process Claims: Evidence from a Century of U.S. Patents," unpublished manuscript available at <a href="https://papers.ssrn.com/abstract=4069994">https://papers.ssrn.com/abstract=4069994</a>.</p>
Classifying protein kinase conformations with machine learning: data
<p>This data collection accompanies the manuscript "Classifying protein kinase conformations with machine learning".</p> <p>It is created using the <a href="https://github.com/edikedik/kinactive">kinactive</a> v0.1 tool written in pure Python v3.10. <strong>Note that the data are provided for the reference and reproducibility purposes and will not be compatible with later versions of `kinactive` built upon <a href="https://github.com/edikedik/lXtractor">lXtractor</a> > 0.1.1.</strong> Refer to the <a href="https://kinactive.readthedocs.io/en/latest/index.html">kinactive documentation</a> for instructions on how to obtain an actualized version of the structural kinome collection.</p> <p>File descriptions:</p> <ul> <li>db_v3.tar.gz -- a structural kinome collection archive. One can unpack it and inspect the contents or load it into the Python interpreter using `kinactive` or `lXtractor` tools.</li> <li>db_af2.tar.gz -- an AlphaFold2 kinome collection for Swiss-Prot sequences.</li> <li>default_*_vs.tsv -- structure/sequence variables calculated with lXtractor and used in an interpretable ML pipeline.</li> <li>*_features.tsv -- lists of ranked features selected by the <a href="https://github.com/edikedik/eBoruta">eBoruta</a> tool for each classifier.</li> <li>Supplement_labels.tsv -- ML model predictions for each PK domain structure found in db_v3.</li> <li>predictions_af2.csv -- Active/Inactive and DFG labels predicted for domains in db_af2.</li> </ul> <p> </p>
16S V4-V5 metabarcoding reference databases and weighted naive-bayes classifiers, dereplicated
<p>16S metabarcoding databases and naive-bayes classifiers specific to the V4-V5 region. Built from the <a href="https://www.arb-silva.de/documentation/release-138/">Silva 138.1 SSU Ref NR 99</a> database using Qiime2 (version 2023.2) and the <a href="https://github.com/BenKaehler/q2-clawback">q2-clawback plugin.</a> Includes weighted classifiers for two Earth Microbiome Project Ontology (EMPO) 3 habitat types: "sediment (saline)" and "water (saline)" , with data downloaded from <a href="https://qiita.ucsd.edu/">Qiita</a>. Sequences were dereplicated with Rescript --p-mode 'uniq' , retaining identical sequence records that have differing taxonomies.</p> <p>Primers used:</p> <p>EMP 16S 515f: GTGYCAGCMGCCGCGGTAA</p> <p>EMP 16S 926r: CCGYCAATTYMTTTRAGTTT</p> <p><strong>Stats</strong></p> <p>286,948 unique sequences</p> <p>309,567 total sequences</p> <p>46,254 unique taxa (Level 7)</p> <table> <caption>File description</caption> <thead> <tr> <th scope="col"> <table> <thead> <tr> <th>File</th> <th>Description</th> </tr> </thead> <tbody> <tr> <td>make new 16S silva V4-V5 database.md</td> <td>Markdown with code used to generate databases</td> </tr> <tr> <td>silva-138-99-seqs.qza</td> <td>Full length Silva 138.1 SSU 99 sequences</td> </tr> <tr> <td>silva-138-99-tax.qza</td> <td>Taxa for full length Silva 138.1 SSU 99 database</td> </tr> <tr> <td>silva-138_1-99-515f_926r-uniq-seqs.qza</td> <td>Sequences for 16S V4-V5 (primers 515f, 926r), extracted from Silva 138.1 SSU 99, generated by qiime2-2023.2 (forward compatible), dereplicated</td> </tr> <tr> <td>silva-138_1-99-515f_926r-uniq-taxa.qza</td> <td>Taxa for silva-138_1-99-515f_926r-seqs.qza database, dereplicated</td> </tr> <tr> <td>uniform-silva-138_1-99-515f_926r-uniq-classifier.qza</td> <td>Unweighted (uniform) naive-bayes classifier for 16S V4-V5 (primers 515f, 926r) extracted from Silva 138.1 SSU 99, generated by qiime2-2023.2 (forward compatible)</td> </tr> <tr> <td>silva-138_1-99-515f_926r-uniq-sediment-saline-classifier.qza</td> <td>Weighted naive-bayes classifier for 16S V4-V5 (primers 515f, 926r) extracted from Silva 138.1 SSU 99, weighted for sediment-saline, generated by qiime2-2023.2 (forward compatible)</td> </tr> <tr> <td>silva-138_1-99-515f_926r-q2_2023_2-uniq-sediment-saline-weights.qza</td> <td>Weights used to generate silva-138_1-99-515f_926r-q2_2023_2-sediment-saline-classifier.qza</td> </tr> <tr> <td>silva-138_1-99-515f_926r-uniq-water-saline-classifier.qza</td> <td>Weighted naive-bayes classifier for 16S V4-V5 (primers 515f, 926r) extracted from Silva 138.1 SSU 99, weighted for water-saline, generated by qiime2-2023.2 (forward compatible)</td> </tr> <tr> <td>silva-138_1-99-515f_926r-uniq-water-saline-weights.qza</td> <td>Weights used to generate silva-138_1-99-515f_926r-water-saline-classifier.qza</td> </tr> </tbody> </table> </th> <th scope="col"> </th> </tr> </thead> <tbody> </tbody> </table> <p> </p>
18S rRNA V9 taxonomic classifier
<p>18S metabarcoding databases and naive-bayes classifier specific to the V9 region. Built from the <a href="https://pr2-database.org/">PR2 v5.0.0 database</a> using Qiime2 (version 2021.2).</p><p>Primers used:</p><p>1391f: GTACACACCGCCCGTC</p><p>EukBr: TGATCCTTCTGCAGGTTCACCTAC</p><p>File Description </p><p>pr2_seqs.qza |. Full length PR2 v5.0.0 sequences</p><p>pr2_tax.qza | Taxa for PR2 v5.0.0 database</p><p>refseqs_V9.qza. | Sequences for 18S V9 (primers 1391f, EukBrr), extracted from PR2 SSU v5.0.0, generated by qiime2-2021.2 (forward compatible)</p><p>classifier_V9.qza | Unweighted (uniform) naive-bayes classifier for 18S V9 extracted from PR2 SSU v5.0.0, generated by qiime2-2021.2 </p><p> </p>
Data from : Classifying wetland‐related land cover types and habitats using fine‐scale lidar metrics derived from country‐wide Airborne Laser Scanning
<p>This data repository contains the processed lidar metrics for characterizing the habitat structure for classifying main land cover and habitat types in the Lauwersmeer area in the northern part of the Netherlands in the province of Groningen (5754 ha). The lidar metrics were derived from Airborne Laser Scanning (ALS) data using the Actueel Hoogtebestand Nederland 2 (AHN2) openly available dataset from https://www.pdok.nl/. </p> <p>The derived lidar metrics saved in *.grd file format and contain 32 bands. Each band represents a lidar metric and the water surface was masked out in the dataset. The *l1* in the file name indicates that the file was used for level 1 (wetland) classification and *l23* used for level 2 (land cover types within wetland) and level 3 (reedbed habitats) classification. The lidar metrics were calculated using lidR (<a href="https://github.com/Jean-Romain/lidR">https://github.com/Jean-Romain/lidR</a>) software package. Further details related to the lidar metrics extraction can be found at <a href="https://github.com/eEcoLiDAR/PhDPaper1_Classifying_wetland_habitats">https://github.com/eEcoLiDAR/PhDPaper1_Classifying_wetland_habitats</a> Github repository.</p> <p> </p>
Residential Building Image Classifier
<p>This model requires TensorFlow 2.3.0 or above.</p> <p> </p> <p>Classes are:</p> <p> </p> <p>0 : multi-family</p> <p>1 : single-family-1-story</p> <p>2 : single-family-2-or-more-story</p>
Classifying hot water chemistry: Application of MULTIVARIATE STATISTICS - Dataset
<p>These files are the dataset for the following paper "Classifying hot water chemistry: Application of MULTIVARIATE STATISTICS". Authors: Prihadi Sumintadireja<sup>1</sup>, Dasapta Erwin Irawan<sup>1</sup>, Yuano Rezky<sup>2</sup>, Prana Ugiana Gio<sup>3, </sup>Anggita Agustin<sup>1</sup></p>
Data and code used in analyses for Simulated soundscapes and transfer learning boost the performance of acoustic classifiers under data scarcity
<p>Evaluation datasets, Python scripts, and computation environments used to conduct analyses for Simulated soundscapes and transfer learning boost the performance of acoustic classifiers under data scarcity. <br><br>transfer_learning_project.zip also contains a vignette describing the use of a generalized script for adapting these methods to novel acoustic classification tasks. </p> <p> </p>
Replidec - Use naive Bayes classifier to identify virus lifecycle from metagenomics data
<p>Replidec: Replication Cycle Decipher for Phages</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.