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.
13
datasets available to search
ShareScore release 0.9.0
Dataset results
13 results for “binary classification”
Data set for (binary) text classification, involving spoken utterances and written text
<p>This data set contains sentences belonging to either of two classes: Transcripts of spoken<br> (informal) text (Class 0), and written, formal text (Class 1). Sentences in Class 0 were<br> obtained from publicly available transcripts of radio shows (e.g. NPR),<br> whereas Sentences in Class 1 were obtained from Wikipedia.</p> <p>The data set is divided into three subsets: Training, validation, and test (specified by the file names).<br> Each set contains a large number of sentences, belonging to either of the two classes:</p> <p>In total, there are 13,640,458 sentences, of which 6,374,487 in Class 0 and 7,265,971.<br> The training set contains 9.743,188 sentences (of which 4,553,205 in Class 0 and 5,189,983 in Class 1), <br> the validation set contains 1,948,639 sentences (of which 910,641 in Class 0 and 1,037,998 in Class 1), and the <br> test set contains 1,948,631 sentences (of which 910,641 in Class0 and 1,037,990 in Class1). </p> <p>The data sets are in plain text format. Every row contains (i) the class label (0 or 1) and<br> (ii) the text of the sentence, separated from the class label by a tab character.</p> <p>Note that the sentences contain 5 tokens or more (including punctuation marks). </p>
Jacobaea vulgaris and meadow image classification dataset (binary)
<h3>General Information</h3> <p>Instances in the Jacobaea vulgaris class: 895<br>Instances in the Meadow class: 9141<br>Image sizes from 77x77 to 817x817 pixels on three color channels (RGB)</p> <p> </p> <h3>Data Generation and Source</h3> <p>The images in this dataset were taken as part of the project “UAV-basiertes Grünlandmonitoring auf Bestands- und Einzelpflanzenebene” (engl. “UAV-based Grassland Monitoring at Population and Individual Plant Level”), financed by the Authority for Economy, Transport, and Innovation of Hamburg. <br>In September 2018, flights with an octocopter were conducted over two extensively used grassland areas in the urban area of Hamburg. The multicopter flew in a height of circa 11 meters and took pictures with a ground resolution of approximately 3,18 mm/pixel. Additional information about the process of image generation for this dataset are to be found in the relevant papers written by P. Zacharias: 1) <a href="https://archiv.geomv.de/geoforum/2019/doc/Tagungsband_GeoForum-MV-2019_eBook.pdf" target="_blank" rel="noopener">UAV-basiertes Grünland-Monitoring und Schadpflanzenkartierung mit offenen Geodaten</a> [p. 45–53] and 2) <a href="https://www.auf.uni-rostock.de/storages/uni-rostock/Alle_AUF/AUF/GG/PDF/gruenlandmonitoring/2019-12-12-FHH-Workshop_Vortrag_Zacharias.pdf" target="_blank" rel="noopener">UAV-basiertes Grünlandmonitoring auf Bestands- und Einzelpflanzenebene</a>.</p> <p>Additionally, to the images of Jacobaea vulgaris taken by the UAV, the dataset includes images of Jacobaea vulgaris plants from the internet (included in the total 895 images; e.g. images 'jkk0523.jpg', 'jkk0527.jpg'). Furthermore, some of the images of the Jacobaea vulgaris plants have been rotated, further cropped or a filter has been applied. The exact number of augmentations made is unknown. As there are augmented images included in the datasets -which makes the dataset useful for training and validation- a use of the dataset for testing purposes is not recommended due to the risk of data leakage.</p> <h3>Data License</h3> <p>The dataset is licensed under the license CC BY 4.0. The attributor of the data is the Chair of Geodesy and Geoinformatics at the University of Rostock. The data was created within the scope of the project 'UAV-based Grassland Monitoring at Population and Individual Plant Level', financed by the Authority for Economy, Transport, and Innovation of Hamburg.</p> <p> </p>
Jacobaea vulgaris and meadow Augmented image classification dataset (binary)
<h3>General Information</h3> <p>Total instances: 117008<br>Instances in the Jacobaea vulgaris class: 58504 <br>Instances in the Meadow class: 58504<br>Image sizes from 224x224 pixels on three color channels (RGB)</p> <p><br>Performance increase training a ResNet50 on the base dataset versus the same architecture on the augmented data set shared here: +3,79 percent points in ROC AUC on an independent test set with 240 instances.<br><br></p> <h3>Data Generation and Source</h3> <p>The initial images in this dataset were taken as part of the project “UAV-basiertes Grünlandmonitoring auf Bestands- und Einzelpflanzenebene” (engl. “UAV-based Grassland Monitoring at Population and Individual Plant Level”), financed by the Authority for Economy, Transport, and Innovation of Hamburg. <br>In September 2018, flights with an octocopter were conducted over two extensively used grassland areas in the urban area of Hamburg.</p> <p>In my master's thesis at <a href="https://www.tu.berlin/dams">DAMS Lab</a> at TU Berlin, I evaluated the effect of different augmentation strategies for Jacobaea vulgaris image classification on the several performance metrics (most importantly the ROC AUC score). The identified augmentation strategies are -besides to performance based selection- also selected based on domain knowledge, which I acquired during the research for my master thesis. </p> <p>Additional information about the initial image generation process is to be found <a href="https://archiv.geomv.de/geoforum/2019/doc/Tagungsband_GeoForum-MV-2019_eBook.pdf">here </a> [p. 45–53] and <a href="https://www.auf.uni-rostock.de/storages/uni-rostock/Alle_AUF/AUF/GG/PDF/gruenlandmonitoring/2019-12-12-FHH-Workshop_Vortrag_Zacharias.pdf">here</a>. </p> <h3> </h3> <h3>Augmentations applied</h3> <ul> <li>Gaussian Noise: For the Gaussian noise augmentation, the mean of the added noise is set to zero. The lower and upper bounds for the random variance of the noise are 20.4663 and 54.0395 respectively. The bounds were identified by hyperparameter tuning. The search space for the lower bound was set from 5 to 30 and for the upper bound from 31 to 100. Those two search spaces were defined by visual inspection of the effects of applying Gaussian noise with different variance values to images of both classes. The Gaussian noise is sampled for each color channel individually. </li> <li>Random Brightness and Contrast: The brightness will randomly be increased or decreased by a factor ranging from 0.7010 to 1.2990. The The contrast will also be randomly increased by a factor ranging from 0.5775 to 1.4225. Those two ranges were identified using hyperparameter tuning. The search space for the maximal percentual increase or decrease of brightness and contrast was individually set from 1% to maximally 50% increase or decrease.</li> <li>Cutout Dropout: In this augmentation method a certain percentage of the input image is getting covered by black patches. The patches have a certain size in pixels, the implementation of this technique in this thesis uses square patches. The black patches are then randomly introduced into the image, by randomly alloacting the<br>patches across the image and then setting the corresponding pixel values to zero. The iamge is getting covered with patches until the cover percentage is reached. We<br>set percentage of the image to be randomly covered by black patches to 56.76%. The size of the patches, which randomly cover the image, is set to 4 pixels. A<br>good illustration of this is found in figure 4.2. The augmentation technique is inspired by the research proposed by Devries et al.[8]. Both values were identified by hyperparameter tuning. The search space for the patch size in pixels is categorical and includes the values [1, 2, 4, 7, 8, 14, 16, 28]. Those values all are multiples of 224, which is the image width and height in pixels. The patch size needs to be a multiple of the width and height in order to be suitable for the algorithm implementation. The search space for the cover percentage of the image had been set from 1% to 60%. This search space limits narrows the search down to a space where still a big part of the image is uncovered. The algorithm rearranges the image into a two dimensional grid and randomly masks rows of this grid by setting the pixel values in this row to zero. Then, the image gets rearranged, now with the randomly generated patches included.</li> <li>Random Saturation: The saturation of each pixel is randomly getting shifted. The upper bound for randomly shifting<span> </span>the saturation value of each pixel is set to 231.689%. This value was identified using hyperparameter tuning. An upper limit of the maximal saturation shift had been set to 40% shift in either direction for hyperparameter tuning.</li> <li>Horizontal Flip: The image gets flipped along the horizontal axis. </li> <li>Vertical Flip: The image gets flipped along the vertical axis.</li> <li>Random Rotation 90 degrees: Randomly rotates the image by a k-fold of 90 degrees, whereby k = {0, 1, 2, 3}.</li> </ul> <p> </p> <p>All augmentation methods and with their tuned augmentation hyperparameters (if existent) are applied to an image from the test set in figure 4.2. With the seven identified<br>augmentation techniques a dataset of 800% the size of the original dataset is created. The Augment model is trained on exactly this dataset. Of course next to the augmented images, the dataset still includes the original, unaugmented images. TensorFlow, along with additional libraries including Optuna for hyperparameter optimization and Albumentations for image augmentation, were used in for the implementation of this project.</p> <p> </p> <h3>Rational behind the augmentations applied</h3> <ul> <li>Random Rotation, Vertical and Horizontal Flip: These three augmentation strategies were chosen to make the classifier less sensitive to the orientation of the plant. The goal is to train a model that can classify plants regardless of their orientation. In order to achieve this effectively across different orientations, vertical flips, horizontal flips, and random 90-degree rotations are chosen for evaluation.</li> <li>Random Saturation: The varying saturation of the images simulates different levels of chlorophyll in the leaves, which is responsible for the green color of the<br>leaves and the intensity of this color. The color of the plant parts (leaves, stems, and flowers) is also influenced by factors such as soil, sun, weed density and pressure, location, and water availability. Varying the saturation of the images simulates changes in these factors.</li> <li>Gaussian Noise: By adding noise, in this case Gaussian noise, different lighting conditions are simulated when capturing the images. We specifically chose Gaussian<br>noise because it is common in many real-world scenarios and is based on the Central Limit Theorem, which states that the sum of many independent random variables.<br>tends to be normally distributed. This makes Gaussian noise a logical choice for simulating real-world random noise.</li> <li>Random Brightness Contrast: The Random Brightness and Random Contrast Augmentation uses brightness to mimic varying lighting conditions and contrast to highlight differences between plants by contrasting them more strongly, thereby highlighting their edges. This approach for highlighting edges is of course much more subtle than the canny edge detection augmentation. This augmentation method combines a weak focus on edges with variations in lighting conditions in one approach. The random contrast is a much softer approach for highlighting edges of plants, compared to the Canny edge detection augmentation. The other features in the images do not get changed that much, compared to the changes from edge detection augmentation.</li> <li>Cutout Dropout: The cutout augmentation simulates random occlusion by other plants. These occlusions are common and expected. Jacobaea vulgaris plants may be partially or completely obscured by other plants during image capturing. This augmentation technique makes the models more robust to random occlusion.</li> </ul> <h3> </h3> <h3>Data License</h3> <p>The dataset is licensed under the license CC BY 4.0. The attributor of the data is the Chair of Geodesy and Geoinformatics at the University of Rostock. The data was created within the scope of the project 'UAV-based Grassland Monitoring at Population and Individual Plant Level', financed by the Authority for Economy, Transport, and Innovation of Hamburg.</p>
Binary classification of the Kinshasa and Bandundu provinces in the Democratic Republic of the Congo — settled versus non settled
<p>This dataset was created based on a settlement layer produced by the Oak Ridge National Laboratory using feature extraction from high-resolution imagery for population modelling work undertaken in the Kinshasa and Kongo-Central provinces in the Democratic Republic of the Congo. The settlement layer consists of settlement polygons of approximately 7 meters resolution. The polygons were rasterized based on a reference grid with a resolution of 3 arc-seconds, approximately 90 meters. The presence of at least one settlement polygon designated a settled cell.</p> <p>We thank the Oak Ridge National Laboratory and the Bill and Melinda Gates Foundation for the support. We would also like to extend our gratitude to Eric M. Webber and Amy N. Rose at the Oak Ridge National Laboratory and Io Blair-Freese at the Bill and Melinda Gates Foundation.</p>
Appendix Figures A.1 - A.15 of the paper "Advanced classification of hot subdwarf binaries using artificial intelligence techniques and Gaia DR3 data". This work has been accepted for publication in the journal Astronomy & Astrphysics (A&A) on September 24, 2024.
<p><strong>Figure captions:</strong></p> <p> </p> <p><strong>Fig. A.1.</strong> Heatmap with the number of common stars (true positives) labeled as binary for the five methods used.</p> <p> </p> <p><strong>Fig. A.2.</strong> Heatmap with the number of common stars (true negatives) labeled as single for the five methods used.</p> <p> </p> <p><strong>Fig. A.3.</strong> Color-magnitude diagrams, showing the 2815 stars of our sample from Sect. 3. Colors indicate the label predictions by SOM (left panel) and CNN (right panel).</p> <p> </p> <p><strong>Fig. A.4.</strong> K-S test comparing radial SOM (black) and CNN (blue).</p> <p> </p> <p><strong>Fig. A.5.</strong> Spectra of the star "LAMOSTJ112914.11+471501.7" (blue color) and in the background (gray color) the 35 stars classified as binary by Solano et al. (2022) with VOSA tools.</p> <p> </p> <p><strong>Fig. A.6.</strong> Spectra of the star "HD14829" (red color) and in the background (gray color) the 53 stars classified as single by Drilling et al. (2013).</p> <p> </p> <p><strong>Fig. A.7.</strong> Spectra of the star "Feige98" (red color) and in the background (gray color) the 53 stars classified as single by Drilling et al. (2013).</p> <p> </p> <p><strong>Fig. A.8.</strong> Spectra of the star "PG0304+184" (red color) and in the background (gray color) the 53 stars classified as single by Drilling et al. (2013).</p> <p> </p> <p><strong>Fig. A.9.</strong> Spectra of the star "PG1510+635" (red color) and in the background (gray color) the 53 stars classified as single by Drilling et al. (2013).</p> <p> </p> <p><strong>Fig. A.10.</strong> Cluster 0 of spectra (blue color) with the other spectra in the background (gray color).</p> <p> </p> <p><strong>Fig. A.11.</strong> Cluster 4 of spectra (brown color) with the other spectra in the background (gray color).</p> <p> </p> <p><strong>Fig. A.12.</strong> Cluster 1 of spectra (yellow color) with the other spectra in the background (gray color).</p> <p> </p> <p><strong>Fig. A.13.</strong> Cluster 3 of spectra (red color) with the other spectra in the background (gray color).</p> <p> </p> <p><strong>Fig. A.14.</strong> Cluster -1 of spectra (pink color) with the other spectra in the background (gray color).</p> <p> </p> <p><strong>Fig. A.15.</strong> Cluster 2 of spectra (green color) with the other spectra in the background (gray color).</p>
Binary Classification as a Phase Separation Process (data repository)
<p><strong>For version 0.0.2 (from 2021) see below:</strong></p> <p>This is a data repository for the paper "Binary classification as a phase separation process", by Rafael Monteiro.</p> <ul> <li>Website with description of this project: <a href="https://rafael-a-monteiro-math.github.io/Binary_classification_phase_separation/index.html">https://rafael-a-monteiro-math.github.io/Binary_classification_phase_separation/index.html</a></li> <li>Github: <a href="https://github.com/rafael-a-monteiro-math/Binary_classification_phase_separation">https://github.com/rafael-a-monteiro-math/Binary_classification_phase_separation</a></li> </ul> <p>This is a second version, which I wrote using tensorflow. It is much smaller (5 Gb when decompressed), a remarkable improvement when compared to the more than 100 Gb of the previous version).</p> <p>The new files are </p> <ul> <li>PSBC_BCs.tar.gz</li> <li>PSBC_classifier_PCA.tar.gz</li> <li>PSBC_dataset.tar.gz</li> <li>PSBC_libs_grids_statistics.tar.gz</li> <li> PSBC_notebooks.tar.gz</li> </ul> <p>Their content is explained in the file README_v2.pdf</p> <p><strong>UPDATE: <a href="https://drive.google.com/drive/folders/18l_92HuHDWJDkZnvXRuyGedcyC_3YZ2M?usp=sharing">a Google Colab folder is also available</a>. You can also find all the data and libraries there, unpacked.</strong></p> <p>For usage, see the Git-hub. </p> <blockquote> <p><strong>NOTE)</strong> I will keep the content for the previous version available in my Github as well. It is still a "nice exercise" to do all that is done in this new version in numpy, as done there. <strong><em>(Or, I should say, they should be studied as a cautionary tale of what to avoid.)</em></strong></p> </blockquote> <p> </p> <p><strong>For version 0.0.1 (from 2020) see below:</strong></p> <p>This is a data repository for the paper "Binary classification as a phase separation process", by Rafael Monteiro.</p> <ul> <li>Website with description of this project: <a href="https://rafael-a-monteiro-math.github.io/Binary_classification_phase_separation/index.html">https://rafael-a-monteiro-math.github.io/Binary_classification_phase_separation/index.html</a></li> <li>Github: <a href="https://github.com/rafael-a-monteiro-math/Binary_classification_phase_separation">https://github.com/rafael-a-monteiro-math/Binary_classification_phase_separation</a></li> </ul> <p>Therein you will find</p> <ul> <li>Examples</li> <li>1D toy model examples</li> <li>Computational statistics</li> <li>Several trained PSBC on MNIST dataset, with different parameter configurations</li> <li>Extra simulations, investigating normalization properties, low dimensional models that fail due to "too much" model compression, and comparison among ANNs, KNNs, and the PSBC in 1D</li> </ul> <p>If you want to know</p> <ol> <li>how to read the data</li> <li>how to access computational statistics, raw data, and examples</li> <li>how to use the data stored in this data repository</li> </ol> <p>see the guide README.pdf on GitHub page at <a href="https://github.com/rafael-a-monteiro-math/Binary_Classification_Phase_Separation">Binary_Classification_Phase_Separation</a>, where a script that downloads (and organizes) all this data is also available ("download_PSBC.sh).</p> <p>I did not include a copy of the train-test set (0-1dubset of the MNIST database) in every folder with simulations. But you can find a copy of the normalized dataset in the tar ball "PSBC_Examples.tar.gz" as</p> <p>data_test_normalized_MNIST.csv and data_train_normalized_MNIST.csv.</p> <p> </p>
Dataset related to the article "Binary classification of copy number alteration profiles in liquid biopsy with potential clinical impact in advanced NSCLC"
<p>This record contains original data used in the article "Binary classification of copy number alteration profiles in liquid biopsy with potential clinical impact in advanced NSCLC" to develop a linear support vector machine (SVM) classifier to predict chromosomal instability. </p> <p>We retrospectively evaluated the results of plasma NGS analysis performed at our Institution by using the AVENIO ctDNA Expanded Kit, a panel of 77 genes, which detects the major classes of genetic alterations. Binary classification, into “stable” (SCP) or “unstable” (UCP) chromosomal profiles, was initially performed by visual inspection of individual CNV alteration profiles by two independent professionals of our group. Then we decided to implement a support vector machine (SVM) classifier to automatically classify CNV profiles as SCP or UCP, beyond operators’ experience. We considered the segmented log2 ratios (.cns) files provided by the CNV kit software and computed three features (Segments, Size, Chromosomes). An alteration (“occurrence of instability”) in the CNV profile was defined each time we found a DNA segment of any size with absolute value of the log2 copy ratio exceeding a fixed cut-off. Two different cut-off values on log2 copy ratio were examined: 0.1 and 0.2. Once the cut-off was defined, three features were considered as covariates in the SVM classifier: 1) number of altered segments (Segments), 2) total length of altered regions (Size) and 3) number of affected chromosomes (Chromosomes).</p> <p>The “dataset_0.1.txt” and “dataset_0.2.txt” files are the original data matrices obtained by considering a cut-off of 0.1 and 0.2, respectively, on the absolute value of the log2 copy ratio.</p> <p>Rows represent available samples in our study (n=177). Columns contain the following variables: anonymized sample IDs (Sample), the class, “stable” or “unstable”, as assigned by two independent professionals of our group (Class), the corresponding binary label (Label: 0 for “stable”, 1 for “unstable”), the three features used as covariates in the SVM classifier and computed as described above (Segments, Size, Chromosomes).</p> <p><br>For the detailed results of our work, please refer to the full article.</p>
Data and code for "Des-q: a quantum algorithm to construct and efficiently retrain decision trees for regression and binary classification"
<p>It contains the code and the data to reproduce the figures in the paper "Des-q: a quantum algorithm to construct and efficiently retrain decision trees for regression and binary classification" published in arXiv: https://arxiv.org/abs/2309.09976</p>
Binary Classification Datasets
Open the record for dataset details and reuse information.
Synthetic Datasets for "Binary Classification Optimisation with AI-Generated Data"
<p>Images of melanomas and Basal Cell Carcinoma generated with a stylegan2. Dataset corresponding to the article "Binary Classification Optimisation with AI-Generated Data"</p>
Data and Software for Astrometric Binary Classification
<p>The following data set and software accompany the paper "Astrometric Binary Classification Via Artificial Neural Networks" by Joe Smith, accepted to the Astrophysical Journal (ApJ) in August 2024. </p> <p>The data and the data mining techniques used to create the data set to train the ANN can be found in Section 2.2 of the paper - this data set is provided below for ease of access. There is also a code file containing the model's weights, architecture, training configuration, and additional information:</p> <p>1. astrobinary_data.csv<br>2. ann_model.keras</p> <p>To make classifications, the ANN can be loaded via the ann_model.keras file with the following Python script:</p> <p>import tensorflow as tf<br>ann_model = tf.keras.models.load_model('ann_model.keras')<br>ann_model.summary()<br><br>DISCLAIMER:<br>If you download and use this ANN model, please cite the paper above.</p>
M 83 Chandra X-Ray Binary Classifications Using HST
Building on recent work by Chandar+ (2020, J/ApJ/890/150), the authors constructed X-ray luminosity functions (XLFs) for different classes of X-ray binary (XRB) donors in the nearby star-forming galaxy M 83. Rather than classifying low-versus high-mass XRBs based on the scaling of the number of X-ray sources with stellar mass and star formation rate, respectively, this catalog utilized multi-band Hubble Space Telescope imaging data to classify each Chandra-detected compact X-ray source as a low-mass (i.e., donor mass < ~3M<sub>sun</sub>), high-mass (donor mass > ~8M<sub>sun</sub>), or intermediate-mass XRB based on either the location of its candidate counterpart on optical color-magnitude diagrams or the age of its host star cluster. In addition to the standard (single and/or truncated) power-law functional shape, the authors approximated the resulting XLFs with a Schechter function. They identified a marginally significant (at the 1-sigma to 2-sigma level) exponential downturn for the high-mass XRB XLF, at l~38.48<sub>-0.33</sub><sup>+0.52</sup> (in log CGS units). In contrast, the low- and intermediate-mass XRB XLFs, as well as the total XLF of M 83, are formally consistent with sampling statistics from a single power law. This method suggests a non-negligible contribution from low- and possibly intermediate-mass XRBs to the total XRB XLF of M 83, i.e., between 20% and 50%, in broad agreement with X-ray-based XLFs. More generally, the authors caution against considerable contamination from X-ray emitting supernova remnants to the published, X-ray-based XLFs of M 83, and possibly all actively star-forming galaxies. This table presents a fully classified catalog of X-ray sources in M 83 that builds upon the deep Chandra ACIS imaging data published in Lehmer+ 2019 (J/ApJS/243/3). Out of a total of 456 point-like sources brighter than 10<sup>35</sup>erg/s, this work restricts the analysis to the 325 objects that fall within the M 83 HST footprint. HST observations of M 83 were taken with the WFC3/UVIS instrument, spanning seven fields that each cover approximately 162" x 162" for a total mosaic area of ~43 arcmin<sup>2</sup>. All observations were obtained between 2009 August and 2012 September by R. O'Connell (Prop ID. 11360) and W. Blair (Prop ID. 12513), with exposure times ranging from ~1.2 to 2.7 ks for each image. Images were downloaded from the Hubble Legacy Archive (HLA). In general, BVI images are created using the F438W, F547M, and F814W filters. The central field, which includes the galaxy nucleus, uses the broader F555W V-band filter, rather than F547M. The authors also use U-band images (F336W) to help calculate cluster ages. This table was created by the HEASARC in March 2023 based upon the <a href="https://cdsarc.cds.unistra.fr/ftp/cats/J/ApJ/912/31">CDS Catalog J/ApJ/912/31</a> file tablea2.dat. This is a service provided by NASA HEASARC .
Binary Field of Research Classification Data
<p>The zip file includes the following data:</p> <ol> <li>Class category list (for categorical baseline),</li> <li>Document text: containing the title + abstract of each dataset instance.</li> <li>Class ORKG texts only: containing the dataset's ORKG label for each instance.</li> <li>Class DBpedia texts only: containing texts extracted from the linked DBpedia entity for each instance (the text is derived from rdfs:label + rdfs:comment)</li> <li>Class KGEs: containing DBpedia entity embeddings for each instance. The embeddings are pre-trained and extracted from https://zenodo.org/records/6384728.</li> <li>Author embedding per instance.</li> <li>Publisher embedding per instance. </li> <li>Labels: containing binary labels indicating whether an instance is a positive or negative sample. </li> </ol>
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.