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.
199
datasets available to search
ShareScore release 0.9.0
Dataset results
199 results for “active learning”
Fink: early supernovae Ia classification using active learning
<p>Data and code used to obtain results presented in <a href="https://arxiv.org/abs/2111.11438">Leoni et al., 2021, Fink: early supernovae Ia classification using active learning</a></p>
Estimating Orchestration Load in Collaborative Learning Situations Using EDA - Activity 6
<p>Skin conductivity of the teacher while orchestrating a Pyramid activity. The green highlight indicates an SCR concurred with the teacher report: "When students told me that after increasing time in the "improving phase" they could not continue editing their improved answer.".</p>
Estimating Orchestration Load in Collaborative Learning Situations Using EDA - Activity 2
<p>Skin conductivity of the teacher while orchestrating a Pyramid activity. During this activity, the teacher reported: "I noticed that the scenario for the task that I shared with student was not the one I planned (I have several ... and was confused with the one I picked) so I had to read the scenario as well .. while students where completing the Pyramid activity. In any case I know all scenarios very well and was quick for me to remember it.".</p>
Estimating Orchestration Load in Collaborative Learning Situations Using EDA - Activity 3
<p>Skin conductivity of the teacher while orchestrating a Pyramid activity. The teacher did not report any stressful situation during this activity.</p>
Estimating Orchestration Load in Collaborative Learning Situations Using EDA - Activity 1
<p>Skin conductivity of the teacher while orchestrating a Pyramid activity. The green highlight indicates an SCR concurred with the teacher report: "I was running out of time, and needed to reduce time in the XXX activity. However I'm used to this kind of situations, and was not highly stressed."</p>
Multi-task self-supervised learning for wearables - human activity recognition
<p>Datasets used to train and evaluated the self-supervised-learning model</p>
Alarm cues and alarmed conspecifics: Neural activity during social learning from different cues in Trinidadian guppies
<p>Learning to respond appropriately to novel dangers is often essential to survival and success, but carries risks. Learning about novel threats from others (social learning) can reduce these risks. Many species, including the Trinidadian guppy (<em>Poecilia reticulata</em>), respond defensively to both conspecific chemical alarm cues and conspecifix anti-predator behaviours, and in other fish such social information can lead to a learned aversion to novel threats. However, relatively little is known about the neural substrates underlying social learning and the degree to which different forms of learning share similar neural mechanisms. Here, we explored the neural substrates mediating social learning of novel threats from two different conspecific cues (i.e. social cue-based threat learning). We first demonstrated that guppies rapidly learn about threats paired with either alarm cues or with conspecific threat responses (demonstration). Then, focusing on acquisition rather than recall, we discovered that phospho-S6 expression, a marker of neural activity, was elevated in guppies during learning from alarm cues in the putative homologue of the mammalian lateral septum and the preoptic area. Surprisingly, these changes in neural activity were not observed in fish learning from conspecific demonstration. Together, these results implicate forebrain areas in social learning about threat but raise the possibility that circuits contribute to such learning in a stimulus-specific manner.</p>
Supplementary Material - Active Learning in Software Engineering A Rapid Review
<p>Supplementary Material for paper Active Learning in Software Engineering: A Rapid Review presented in CLEI 2024. The spreadsheet contains the studies considered for the Rapid Review.<br><br></p>
Sentinel-2 reference cloud masks generated by an active learning method
<p> <strong>Reference classifications generated with Active Learning for Cloud Detection (ALCD)</strong></p> <p>This data set provides a reference cloud mask data set for 38 Sentinel-2 scenes. These reference masks have been created with the ALCD tool, developed by Louis Baetens, under the direction of Olivier Hagolle at CESBIO/CNES[1]. They were created to validate the cloud masks generated by the MAJA software [2].</p> <p>- The `Reference_dataset` directory contains 31 scenes selected in 2017 or 2018.<br> - The `Hollstein` directory contains 7 scenes that were used to validate the ALCD tool by comparison to manually generated reference images kindlyprovided by Hollstein et al[3]<br> One of these scenes is present in both directories. For the validation of MAJA, the "Hollstein" scenes were not used because of their acquisition at a time period when Sentinel-2 was not yet operational, with a degraded repetitivity of observations.</p> <p><strong># Description of the data structure</strong><br> The name of each scene directory is the name of the corresponding Sentinel-2 L1C product.<br> In the scene directory, three sub-directories can be found.<br> - `Classification`<br> - `Samples`<br> - `Statistics`</p> <p><strong># Description of the files</strong><br> - `Classification/classification_map.tif` --- the main product, which is the classified scene. 7 classes are available. Each one is represented with a different integer.<br> 0: no_data.<br> 1: not used.<br> 2: low clouds.<br> 3: high clouds.<br> 4: clouds shadows.<br> 5: land.<br> 6: water.<br> 7: snow.</p> <p>- `Classification/confidence_enhanced.tif` --- enhanced confidence map of the classification. The values are between 0 and 255 (coded on 1 bit).<br> The original confidence map is, for each pixel, the proportion of votes for the majority class as the classification map has been created via a Random Forest algorithm.<br> A median filter has been applied to this confidence map. Finally, the value was saved on 1 bit, leading to the value being between 0 and 255.</p> <p>- `Classification/contours.png` --- the contours of the classes from the classification map, overlayed on the scene. The color code depends on each class.<br> Green: low and high clouds. Yellow: cloud shadows. Blue: water. Purple: snow.</p> <p>- `Classification/used_parameters.json` --- the parameters that were used to classify the scene. It includes the tile code, the cloudy and clear dates, along with their product reference.</p> <p>- `Samples/` --- this directory contains all the shapefiles, one per class.</p> <p>- `Statistics/k_fold_summary.json` --- results of the 10-fold cross-validation on the scene.<br> 5 metrics are computed, in the order given in the "metrics_names". "all_metrics" is a list of the 10 folds, with the 5 metrics in the correct order for each fold.<br> "means" and "stds" are the means and standard deviations of the 10 folds.</p> <p><br> <strong># References</strong></p> <p>[1] Baetens, L.; Desjardins, C.; Hagolle, O. Validation of Copernicus Sentinel-2 Cloud Masks Obtained from MAJA, Sen2Cor, and FMask Processors Using Reference Cloud Masks Generated with a Supervised Active Learning Procedure. <em>Remote Sens.</em> <strong>2019</strong>, <em>11</em>, 433.</p> <p>[2] A multi-temporal method for cloud detection, applied to FORMOSAT-2, VENµS, LANDSAT and SENTINEL-2 images, O Hagolle, M Huc, D. Villa Pascual, G Dedieu, Remote Sensing of Environment 114 (8), 1747-1755, 2010</p> <p>[3] Hollstein, A.; Segl, K.; Guanter, L.; Brell, M.; Enesco, M. Ready-to-Use Methods for the Detection of Clouds, Cirrus, Snow, Shadow, Water and Clear Sky Pixels in Sentinel-2 MSI Images. Remote Sens. 2016, 8, 666</p>
Interesting spectra discovered by Active Deep Learning in LAMOST DR2
<p>PDF previews of interesting spectra found by Active Deep Learning in LAMOST DR2 spectra survey. These are referred to in an article Škoda,Podsztavek&Tvrdík "Active deep learning method for the discovery of objects of interest in large spectroscopic surveys" submitted to Astronomy and Astrophysics. Previews show both the whole LAMOST spectrum created from original FITS file and its zoomed part around Halpha line used for machine learning.</p>
Compound activity classes from ChEMBL for machine learning analysis
<p>Ten activity classes are provided that were extracted from ChEMBL version 24 for machine learning studies. Compounds are given in SMILES representations. The following selection criteria were applied. Compounds were required to be tested in a direct binding assay against a single human protein with a ChEMBL assay confidence score of 9. In addition, K<sub>i</sub> measurements had to be available. If multiple K<sub>i</sub> values were available for a compound and did not fall within the same order of magnitude, the compound was not selected. Furthermore only compounds with (mean) pK<sub>i</sub> of at least 5 were considered. Moreover, activity classes had to contain at least 200 compounds belonging to at least 50 computationally determined analog series. The 10 deposited classes consist of 243 to 955 compounds and 57 to 216 analog series.</p>
Traversing Chemical Space with Active Deep Learning for Low-data Drug Discovery
<p>Raw and processed data from LitPCBA used in the paper "Traversing Chemical Space with Active Deep Learning for Low-data Drug Discovery"</p>
Activation and connectivity maps - A chronometric relationship between circuits underlying learning and error monitoring in the basal ganglia and salience network
<p>Activation and connectivity maps of the study "A chronometric relationship between circuits underlying learning and error monitoring in the basal ganglia and salience network".</p> <ul> <li>Error-correct.nii corresponds to the statistical map of group-level differences in BOLD signal between correct and erroneous responses shown in figure 4;</li> <li>Late-initial.nii corresponds to the statistical map of group-level differences in BOLD signal between the initial and late learning periods shown in figure 5;</li> <li>Conn_error-correct_dACC.nii corresponds to the results from the seed-to-voxel gPPI analysis, using the dACC as seed region, showing areas of higher functional connectivity in erroneous compared to correct responses, shown in figure 8.</li> </ul>
Data for Exploring Inhomogeneous Surfaces: Ti-rich SrTiO3(110) Reconstructions via Active Learning
<p>The archive "supplementary_data.tar.gz" contains structures, training data, training scripts, models and an evolution script obtained and reported in the study:<br>"Exploring Inhomogeneous Surfaces: Ti-rich SrTiO3(110) Reconstructions via Active Learning".</p> <p>See README for more information on the archive content.</p>
Out-of-Distribution Evaluation of Active Learning Pipelines for Molecular Property Prediction Dataset
<p>Data for the paper "Out-of-Distribution Evaluation of Active Learning Pipelines for Molecular Property Prediction"</p>
MFB Liposome active learning
Open the record for dataset details and reuse information.
Dataset for "Comparative Analysis of Machine Learning Models to Forecast Flaring Capability of Solar Active Regions: A Parameter Based Approach"
<p>This CSV file contains the values of 14 selected magnetic features along with the active region class for all the regions used in our study. As discussed in the paper, these 14 magnetic features characterize the properties of active regions. All these magnetic features are obtained from the HMI SHARP data series which provides open-sourced vector magnetic field information of solar active regions. the column named 'AR_class' carries information about the class of active regions i.e., 1 for flaring regions and 0 for non-flaring regions.</p>
shemakes activities: record of activities and participation in the learning paths and innovation services.
<p><strong>Shemakes record of activities and participation dataset</strong> contains the number of participants for each event or activity during the project duration collected mainly from the WP2 learning paths and WP3 innovation services, which are the WPs with a high amount of activities and events related to the shemakes development. It does not contain personal or sensitive data, as it is focused on the event's typology. Participants are aggregated per activity. Besides, the dataset includes information about 1) the country where the activity is carried out and 2) it mentions the type of activity and its title. The data about the participants comprises the number of females, whether the females were adults or minors and the percentage of female participants.</p>
Data for: The Neonicotinoid Imidacloprid Impairs Sucrose Solution Consumption, Learning and Locomotor Activity Levels In Bumblebees (Bombus Terrestris)
<p># README</p> <p>The following files are for creating the figures from the paper: </p> <p>## `plot_flowervisits_nectar.ipynb`</p> <p>Jupyter notebook that creates the figures concerning flower visits, nectar consumption and the proportion of empty honeypots.</p> <p>## `plot_activity.py`</p> <p>Python script that takes trajectory fragments from video analysis and computes the locomotor activity level through making histograms of bumblebee speeds. Makes two figures that are equivalent to the figure on locomotor activity in the paper.</p> <p>## `statistical analysis.py`</p> <p>R markdown notebook that performs all the hypothesis testing for the paper.</p> <p>## Data</p> <p>These files contain the data, and are located in the folder called `data`. </p> <p>`activity/activityproportions.csv` contains the computed locomotor activity level for easy plotting.</p> <p>`boldata/boldata.csv` contains data about the nectar bag weight before and after experiment and the counted number of empty and full honeypots. Used by `plot_flowervisits_nectar.ipynb`</p> <p>`flower_data/flowerData.csv` contains the computed number of visits to blue and yellow flowers per hive for easy plotting.</p> <p>`humlevideo_production/*/traj*_trajectories*.csv` contains constructed trajectories from all experiments seen from both cameras. These are being used by the script `plot_activity`.</p> <p>`humlevideo_production/*/traj*.json` contains data about the occurence of bees on flowers in each frame in each experiment.</p> <p>`landinger_csv` contains data about landings, that have been extracted from the `humlevideo_production/*/traj*.json` files. Used by `plot_flowervisits_nectar.ipynb`.</p> <p> </p>
NEATmap: a high-efficiency deep learning approach for whole mouse brain neuronal activity trace mapping
<p>Here are some demo datasets for validating the NEATmap pipeline for high-efficiency whole brain c-Fos<sup>+</sup> cell automated segmentation and quantitative analysis, including:</p> <ol> <li>BrainImage_group.zip.001-007: Validation of NEATmap for automated segmentation and quantitative analysis of mouse whole-brain c-Fos activity images (in Forced Swimming Test).</li> <li>Segmentation_result.zip: Figure 1a, Supplementary Videos 1 and 2 show dual-channel brain slices and segmentation results. They can be merged using Imaris to validate the segmentation results of NEATmap.</li> <li>RawImage_example.zip: High-resolution 3D images of mouse brain slices showing c-Fos<sup>+</sup> cells in Figure 1e.</li> </ol> <p>Due to the total size of the mouse whole-brain image datasets (both raw and processed) included in all the tests exceeding 10 Terabytes, uploading it to a public data repository is impractical. In this work, we provide a dataset of dual-channel (c-Fos<sup>+</sup> channel and autofluorescence channel in forced swimming test experimental group) whole-brain images of mouse for the validation of NEATmap automated segmentation method.</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.