Find research datasets worth reusing

opencc-by-4.0Mar 2023View details →

Investigating Online Art Search through Quantitative Behavioral Data and Machine Learning Techniques - Dataset

This dataset includes the detailed values and scripts used to study behavioral aspects of users searching online for Art and Culture by analyzing quantitative data collected by the Art Boulevard search engine using machine learning techniques. This dataset is part of the core methodology, results and discussion sections of the research paper entitled "Investigating Online Art Search through Quantitative Behavioral Data and Machine Learning Techniques"

opencc-by-4.0Apr 2023View details →

Towards Generalization of Machine Learning Models: An Arabic Sentiment Analysis Dataset

This data set consists of approximately 1.64 Million Arabic tweets (shared by their IDs) posted from 2009 to 2020, and their corresponding sentiment using a three-point classification system of Positive, Negative and Neutral/Mixed. No specific locations and/or keywords were specified throughout the data collection to obtain variation in the dialects and topics represented within the dataset. It is important to note that any biases in the proposed dataset in relation to the dialects and/or topics discussed were unintentional. Please use the following citation if you use this data in a paper: <blockquote> Abdaljalil, S., Hassanein, S., Mubarak, H., & Abdelali, A. (2023). Towards Generalization of Machine Learning Models: A Case Study of Arabic Sentiment Analysis. Proceedings of the International AAAI Conference on Web and Social Media, 17(1), 971-980. </blockquote>

opencc-by-4.0May 2023View details →

Large-scale Docking Datasets for Machine Learning

Large-scale virtual screening has become a valuable tool for early-phase drug discovery. Recent expansions of commercial chemical space have made it computationally intractable to evaluate all compounds in the libraries. Machine learning is one of the methods that aim to prioritize specific subsets of these vast libraries. In order to put these methods to the test, access to large-scale datasets is beneficial. To help the community benchmark their work, we share the docking scores of several ultralarge virtual screening campaigns. The datasets we provide contain canonical SMILES, compound identifiers, and docking scores. We docked two different chemical libraries against eight different biological targets with therapeutic relevance. The first dataset contained approximately 15.5 million molecules adhering to the "Rule-of-Four", whereas the second datasets consists of approximately 235 million "lead-like" molecules. The biological targets represent different classes of proteins and binding sites. More details on the datasets and our methods can be found on (<a href="http://github.com/carlssonlab/conformalpredictor">https://github.com/carlssonlab/conformalpredictor</a>) and our pre-print (<a href="https://doi.org/10.26434/chemrxiv-2023-w3x36">https://doi.org/10.26434/chemrxiv-2023-w3x36</a>). Please feel free to download and use these datasets for your own research purposes. We only ask that you cite our pre-print and datasets appropriately if you use it in your work. Thank you for your interest in our research!

opencc-by-4.0May 2023View details →

Dataset for "Brain-machine interface learning is facilitated by specific patterning of distributed cortical feedback"

Dataset for the manuscript entitled: Brain-machine interface learning is facilitated by specific patterning of distributed cortical feedback. DOI of the manuscript: <a href="https://doi.org/10.1126/sciadv.adh1328">10.1126/sciadv.adh1328</a> Abstract of the manuscript: Neuroprosthetics offer great hope for motor-impaired patients. One obstacle is that fine motor control requires near-instantaneous, rich somatosensory feedback. Such distributed feedback may be recreated in a brain-machine interface using distributed artificial stimulation across the cortical surface. Here, we hypothesized that neuronal stimulation must be contiguous in its spatiotemporal dynamics in order to be efficiently integrated by sensorimotor circuits. Using a closed-loop brain-machine interface, we trained head-fixed mice to control a virtual cursor by modulating the activity of motor cortex neurons. We provided artificial feedback in real time with distributed optogenetic stimulation patterns in the primary somatosensory cortex. Mice developed a specific motor strategy and succeeded to learn the task only when the optogenetic feedback pattern was spatially and temporally contiguous while it moved across the topography of the somatosensory cortex. These results reveal new properties of sensorimotor cortical integration and set new constraints on the design of neuroprosthetics. Description of the variables in the data storage dictionary: <ol> <li>cursor_positions: sequence of the virtual cursor position over a session</li> <li>range_of_rewardable_cursor_position: range of cursor positions that can be rewarded. Upper threshold excluded. Lower threshold included.  </li> <li>cursor_times: timing of the cursor positions provided by the cursor_positions data, in the same clock as spike and lick times. </li> <li>lick_times: timing of all recorded licks.  </li> <li>reward_times:timing of the opening of the valve that releases the water reward. </li> <li>master_spike_times: time of the spikes of the master neurons.</li> <li>master_spike_shape: spike shape of each spike stored in master_spike_times. The shipe shapes are shown for 3s (30kHz sampling rate), for each 4 electrode of the corresponding tetrode.</li> <li>neighbor_spike_times': same as master_spike_times for neighbor neurons.</li> <li>neighbor_spike_shape': same as master_spike_shape for neighbor neurons.   </li> </ol> General structure of the data set: Each variable is a hierarchical tree of lists: [Protocol][Mouse][Session].  Protocol takes one of the following values: 0: Bar feedback | 1: Full shuffle | 2: No Feedback | 5: Playback Structured Feedback | 7: Spontaneous activity | 8: Barrel Shuffle | 9: Frame shuffle. Mouse and Session iterate through respectively the mice that were involved in the protocole, and the sessions, generally 5 in total. Python code to load the hdf5 File The code below relies on libraries available on a standard, mac anaconda install of the jupyter notebook system on the 25/03/2024. It provides with several nested lists of Numpy arrays. For instance, to access the series of cursor positions of Mouse M during session number S of Protocole P, index as follows:  CP = cursor_positions_DATA[P][M][S] # 1 - Load the libraries     import h5py     import numpy as np     import pylab as pl     filename = "./data.h5"     f = h5py.File(filename, "r") # 2 - Extraction of the cursor position, time, lick and reward time, as well as the activity of the master and neighbor neurons.      cursor_positions = f['cursor_positions']     cursor_positions_DATA = []     for Protocole in range(9):         DATA_protocole = []         P = f[cursor_positions[Protocole][0]]         for Mouse in range(16):                 # Loop through the mice, and collect              M = f[P[Mouse][0]]             if not(list(M) == [0,1]) and (len(M) == 5):                 DATA_mouse = []                 for Session in range(5):                     S = f[M[Session][0]]                     bfr = np.array(S)                     DATA_mouse.append(bfr)                  DATA_protocole.append(DATA_mouse)         cursor_positions_DATA.append(DATA_protocole)       cursor_times = f['cursor_times']     cursor_times_DATA = []     for Protocole in range(9):         DATA_protocole = []         P = f[cursor_times[Protocole][0]]         for Mouse in range(16):                 # Loop through the mice, and collect              M = f[P[Mouse][0]]             if not(list(M) == [0,1]) and (len(M) == 5):                 DATA_mouse = []                 for Session in range(5):                     S = f[M[Session][0]]                     bfr = np.array(S)                     DATA_mouse.append(bfr)                  DATA_protocole.append(DATA_mouse)         cursor_times_DATA.append(DATA_protocole)       lick_times = f['lick_times']     lick_times_DATA = []     for Protocole in range(9):         DATA_protocole = []         P = f[lick_times[Protocole][0]]         for Mouse in range(16):                 # Loop through the mice, and collect              M = f[P[Mouse][0]]             if not(list(M) == [0,1]) and (len(M) == 5):                 DATA_mouse = []                 for Session in range(5):                     S = f[M[Session][0]]                     bfr = np.array(S)                     DATA_mouse.append(bfr)                  DATA_protocole.append(DATA_mouse)         lick_times_DATA.append(DATA_protocole)       reward_times = f['reward_times']     reward_times_DATA = []     for Protocole in range(9):         DATA_protocole = []         P = f[reward_times[Protocole][0]]         for Mouse in range(16):                 # Loop through the mice, and collect              M = f[P[Mouse][0]]             if not(list(M) == [0,1]) and (len(M) == 5):                 DATA_mouse = []                 for Session in range(5):                     S = f[M[Session][0]]                     bfr = np.array(S)                     DATA_mouse.append(bfr)                  DATA_protocole.append(DATA_mouse)         reward_times_DATA.append(DATA_protocole)       master_spike_times = f['master_spike_times']     master_spike_times_DATA = []     for Protocole in range(9):         DATA_protocole = []         P = f[master_spike_times[Protocole][0]]         for Mouse in range(16):                 # Loop through the mice, and collect              M = f[P[Mouse][0]]             if not(list(M) == [0,1]) and (len(M) == 5):                 DATA_mouse = []                 for Session in range(5):                     S = f[M[Session][0]]                     DATA_unit = []                     for Unit in range(len(S)):                         if not(list(S) == [0,1]):                             N = f[S[Unit][0]]                             bfr = np.array(N)                             DATA_unit.append(bfr)                      DATA_mouse.append(DATA_unit)                 DATA_protocole.append(DATA_mouse)         master_spike_times_DATA.append(DATA_protocole)       neighbor_spike_times = f['neighbor_spike_times']     neighbor_spike_times_DATA = []     for Protocole in range(9):         DATA_protocole = []         P = f[neighbor_spike_times[Protocole][0]]         for Mouse in range(16):                 # Loop through the mice, and collect              M = f[P[Mouse][0]]             if not(list(M) == [0,1]) and (len(M) == 5):                 DATA_mouse = []                 for Session in range(5):                     S = f[M[Session][0]]                     DATA_unit = []                     for Unit in range(len(S)):                         if not(list(S) == [0,1]):                             N = f[S[Unit][0]]                             bfr = np.array(N)                             DATA_unit.append(bfr)                      DATA_mouse.append(DATA_unit)                 DATA_protocole.append(DATA_mouse)         neighbor_spike_times_DATA.append(DATA_protocole)

opencc-by-4.0May 2023View details →

A Tagged Traffic Accident Dataset for Machine Learning

This dataset contains tagged accident data and is provided for reproducibility for our journal paper  Pablo Moriano, Andy Berres, Haowen Xu, Jibonananda Sanyal. “Spatiotemporal Features of Traffic Help Reduce Automatic Accident Detection Time.” Expert Systems with Applications 244 (2024): 122813. <a title="Persistent link using digital object identifier" href="https://doi.org/10.1016/j.eswa.2023.122813" target="_blank" rel="noopener">https://doi.org/10.1016/j.eswa.2023.122813</a> The accompanying Data in Brief publication discusses the methodology behind the creation of these data. Berres, Andy, Pablo Moriano, Haowen Xu, Sarah Tennille, Lee Smith, Jonathan Storey, and Jibonananda Sanyal. "A Traffic Accident Dataset for Chattanooga, Tennessee." Data in Brief (2024): 110675.   The zip folder annotatedData.zip contains two subfolders: allData and bestData. The bestData folder contains all data for which a full neighborhood of five sensors upstream and five sensors downstream is available, whereas allData includes everything from bestData as well as data with a smaller number of neighboring sensors. Each folder contains one subfolder called accidents and one subfolder called non-accidents. The accidents folder contains one file per accident. The non-accidents folder contains files for the same location, day of the week and time as a corresponding accident, for each week during which there was no accident impact on the traffic. The file names in both folders are formatted as follows: yyyy-mm-dd-hhmm-rrrrrXaaa.a.csv, consisting of date (yyyy-mm-dd), time (hhmm in 24-hour format), and sensor name (rrrrrXaaa.a), which consists of road name (rrrrr; 5 alphanumerical characters), heading (X), and mile marker (aaa.a). For example, the file 2020-11-03-1611-00I24W182.8.csv  contains data for an accident which occurred at 4:11 p.m. on November 3, 2020 on I-24 Westbound near the radar sensor at mile marker 182.8. The content of each CSV file is a timeseries of radar data beginning 15 minutes prior to the reported incident and ending 15 minutes after the reported incident. It also contains metadata, such as the accident type, etc. Each CSV file contains the following columns: <ul> <li>incident at sensor(i): 1 for yes (accidents folder), 0 for no (non-accidents folder)</li> <li>road: road name with heading, e.g. 00I24E</li> <li>mile: mile marker of nearest radar sensor, e.g. 182.8</li> <li>type: accident type, e.g. “Prop Damage (over)” for property damage exceeding a certain threshold. For non-accidents, the type is given as “None”.</li> <li>date: date of the data sample. For accidents, this is the date on which the accident occurred. For non-accidents, this is the date for which the non-accident data sample is collected.</li> <li>incident_time: time the reference accident was reported in hh:mm. This is the time which is provided in E-TRIMS as the time the 911 call was made.</li> <li>incident_hour: just the hour from the incident_time, in integer format.</li> <li>data_time: timestamp for the timeseries contained in the file in hh:mm:ss format. The timeseries consists of 30 second timesteps.</li> <li>weather: weather during data_time, based on data collected from NASA POWER. We used dry bulb temperature (°C), precipitation (mm/h), and wind speed (m/s) from the raw NASA POWER data to produce the classifications of rain (at least 1mm precipitation and temperatures above 2°C), snow (at least 1mm precipitation and temperatures at or below 2°C), and wind (wind speeds over 30 mph or 13.5 m/s). If there were no inclement weather conditions, we set the category to “--".</li> <li>light: light conditions during data_time. To produce this field, we collected sunrise, sunset, civil twilight start and civil twilight end times from <a href="https://sunrise-sunset.org">https://sunrise-sunset.org</a>, and derived the categories dawn, daylight, dusk, and dark using these start and end times.</li> <li>The last 33 columns contain radar data for the 11 sensors surrounding the accident or non-accident. For each sensor, we collected speed (mean over 30-second interval in miles per hour, or empty if no vehicles passed), volume (count of all vehicles passing during 30-second interval), and occupancy (mean % of occupancy over 30-second interval).  These three variables are grouped in triples, of speed (k), volume (k), occupancy (k), where k indicates the sensor number relative to the closest sensor i to the incident, k<i indicate upstream sensors and k>i indicate downstream sensors. For example, speed (i-5) refers to the mean speed at the sensor which is 5 hops upstream from the accident, and volume(i+1) refers to the number of vehicles at the sensor immediately downstream from the accident.</li> </ul> The folder metaData.zip contains the following files: <ul> <li>Accidents.csv: cleaned-up accidents file with all accidents which happened on Chattanooga area highways between November 1, 2020 and April 29, 2021. We have removed accidents which happened on non-highway roads, and we have corrected the timestamps (which were in 12-hour format but missing a.m./p.m. markers) by cross-referencing light and weather conditions.</li> <li>WeatherDict.json: a dictionary containing the weather data synthesized from NASA POWER.</li> <li>LightDict.json: a dictionary containing the light data synthesized from Sunrise-and-Sunset.</li> <li>SensorTopology.csv: neighborhood information for each radar sensor in the Chattanooga area.</li> <li>SensorZones.geojson: polygons used to determine the nearest radar sensor for each accident location. Each polygon is tagged with the corresponding radar sensor’s name.</li> </ul>

opencc-by-nc-sa-4.0Jun 2023View details →

Hainan gibbon (Nomascus hainanus) bioacoustics dataset for machine learning

Data accompanying the paper: "Empowering Deep Learning Acoustic Classifiers with Human-like Ability to Utilize Contextual Information for Wildlife Monitoring" We provide the audio data (.wav) used to test our neural network classifier along with the corresponding labelled text files (.svl). The audio and labelled files can easily be viewed in Sonic Visualiser. Drag and drop the audio file. Create the spectrogram layer. Drag and drop the corresponding .svl file. The dataset provided here is a subset of the full dataset provided here: 10.5281/zenodo.3991714. This dataset has additional files that were manually annotated, which were not manually verified in the original version (10.5281/zenodo.3991714). Files provided <ul> <li>Audio_x.zip -- we provide x number of .zip files containing audio files, numbered 1 to 4. These were created in batches to simplify downloads.</li> <li>Annotations.zip -- .svl files which contain the manually verified labels. These files can be read in Sonic Visualiser, or as .XML files in a programming language. We took care to annotate the start and stop time of each gibbon call. The height of each bounding box is not important as the frequency range of Hainan gibbons is already known.</li> <li>model_weights_tensorflow.hdf5 -- the Tensorflow model. Load the model using: model = tf.keras.models.load_model(model_filepath) note that the model expects a three channel input as explained in the research article.</li> </ul>

opencc-by-4.0Apr 2023View details →

Automatic Identification of Kidney Cell Types in scRNA-seq and snRNA-seq Data Using Machine Learning Algorithms - Datasets

Datasets for reproducibility of the results found in Automatic Identification of Kidney Cell Types in scRNA-seq and snRNA-seq Data Using Machine Learning Algorithms. This study utilized data from the following 4 journals: Lake, B.B. et al. A single-nucleus RNA-sequencing pipeline to decipher the molecular anatomy and pathophysiology of human kidneys. Nat Commun 10, 2832 (2019). Liao, J., Yu, Z., Chen, Y. et al. Single-cell RNA sequencing of human kidney. Sci Data 7, 4 (2020). Menon, R. et al. Single cell transcriptomics identifies focal segmental glomerulosclerosis remission endothelial biomarker. JCI Insight 5, e133267 (2020). Wu, H. et al. Single-cell transcriptomics of a human kidney allograft biopsy specimen defines a diverse inflammatory response. J Am Soc Nephrol 29: 2069–2080 (2018). Young, M. D. et al. Single-cell transcriptomes from human kidneys reveal the cellular identity of renal tumors. Science 361, 594–599 (2018).

opencc-by-4.0Sep 2023View details →

Soil texture dataset from the publication: "Machine learning applied for Antarctic soil mapping: Spatial prediction of soil texture for Maritime Antarctica and Northern Antarctic Peninsula'

Clay, silt and sand distribution in Antarctic soils modeled and predicted through Machine Learning approaches, legacy soil data and environmental covariates. The coefficient of variation and quantile data represent the spatial uncertainty of the predictions. For more information about the methodology used, users are referred to the article:  Siqueira, R.G., Moquedace, C.M., Francelino, M.R., Schaefer, C.E.G.R., Fernandes-Filho, E.I., 2023. Machine learning applied for Antarctic soil mapping: Spatial prediction of soil texture for Maritime Antarctica and Northern Antarctic Peninsula. Geoderma 432, 116405. https://doi.org/10.1016/j.geoderma.2023.116405 The .zip file has the following folders: 1) soil_texture_antarctica: soil texture information containing clay, silt and sand contents 2) soil_texture_coefficient_variation: uncertainty from the coefficient of variation of the soil texture prediction 3) soil_texture_prediction_interval: uncertainty from the prediction interval 90% (Q95% - Q5%) of the soil texture prediction 4) soil_texture_quantile05: quantile 5% of the soil texture prediction 5) soil_texture_quantile95: quantile 95% of the soil texture prediction

opencc-by-4.0Sep 2023View details →

Soil chemistry dataset from the work "Modelling and prediction of major soil chemical properties with Random Forest: machine learning as tool to understand soil-environment relationships in Antarctica"

Bases sum, H+Al (potential acidity), pH, phosphorous, remaining P (P-rem), sodium and total organic carbon distribution in Antarctic soils modeled and predicted through Machine Learning approaches, legacy soil data and environmental covariates. The quantile and prediction interval data represent the spatial uncertainty of the predictions. As soon as the work "Modelling and prediction of major soil chemical properties with Random Forest: machine learning as tool to understand soil-environment relationships in Antarctica" is published, the paper will be cited here.  The .zip file contains the following folders: 1) soil_chemistry_antarctica: data containing the soil chemical attributes distribution 2) soil_chemistry_prediction_interval: uncertainty from the prediction interval 90% (Q95% - Q5%) of the soil attributes prediction 4) soil_texture_quantile05: quantile 5% of the soil attributes prediction 5) soil_texture_quantile95: quantile 95% of the soil attributes prediction

opencc-by-4.0Sep 2023View details →

Image for the dataset in "Extraction of stratigraphic exposures on visible images using a supervised machine learning technique"

This is the original and hand-masked image for the dataset used in a research paper "Extraction of stratigraphic exposures on visible images using a supervised machine learning technique". The content is <ul> <li>Original images with hand-masked images (original_images_NOGUCHIandShoji.zip) <ul> <li>training/* : original images used for the training dataset generation (60 files)</li> <li>training_masks/* : hand-masked images for the training dataset generation (60 files)</li> <li>validation/* : original images used for the training dataset generation (10 files)</li> <li>validation_masks/* : hand-masked images for the validation dataset generation (10 files)</li> <li>test/* : original images used as the test data (5 files)</li> <li>test_masks/* : hand-masked images used as the test data (5 files).</li> </ul> </li> </ul> Note that original images include images obtained using google-image-download, a Python script published on GitHub (<a href="https://github.com/Joeclinton1/google-images-download/tree/patch-1">https://github.com/Joeclinton1/google-images-download/tree/patch-1</a>, Copyright © 2015-2019 Hardik Vasa). The whole images we obtained by google-image-download were labeled as noncommercial reuse with modification. For more details, please refer to a research paper "Extraction of stratigraphic exposures on visible images using a supervised machine learning technique". Correspondence: Rina Noguchi (r-noguchi@env.sc.niigata-u.ac.jp)

opencc-by-4.0Oct 2023View details →

Simulated datasets for detector and particle flow reconstruction: CLIC detector, hit-based data, machine learning format

Derived from https://zenodo.org/record/8260741, prepared in a machine-learning friendly TFDS format, ready to be used with https://zenodo.org/record/8397954. <ul> <li>clic_edm_ttbar_hits_pf10k.tar: ee -> ttbar, center of mass energy at 380 GeV, 10k events</li> <li>clic_edm_qq_hits_pf10k.tar: ee -> Z* -> qqbar, center of mass energy at 380 GeV, 10k events</li> </ul> Contents Each .tar file contains the dataset in the <a href="https://github.com/tensorflow/datasets">tensorflow-datasets</a> (minimum version v4.9.1), <a href="https://github.com/google/array_record">array_record</a> format. Dataset semantics Each dataset consists of events that can be iterated over using the tensorflow-datasets library in either tensorflow or pytorch. Each event has the following information available: <ul> <li>X: the reconstruction input features, i.e. tracks and calorimeter hits</li> <li>ygen: the ground truth particles with the features ["PDG", "charge", "pt", "eta", "sin_phi", "cos_phi", "energy", "jet_idx"], with "jet_idx" corresponding to the gen-jet assignment of this particle</li> <li>ycand: the baseline Pandora PF particles with the features ["PDG", "charge", "pt", "eta", "sin_phi", "cos_phi", "energy", "jet_idx"], with "jet_idx" corresponding to the gen-jet assignment of this particle</li> </ul> The full semantics, including the list of features for X, are available at https://github.com/jpata/particleflow/blob/v1.6/mlpf/heptfds/clic_pf_edm4hep_hits/utils_edm.py.

opencc-by-4.0Mar 2020View details →

Image Dataset for 'Digitally deconstructing leaves in 3D using X-ray microcomputed tomography and machine learning'

Dataset used in the manuscript 'Digitally Deconstructing Leaves in 3D Using X-ray microcomputed Tomography and Machine Learning'. Please cite the paper presenting this dataset: Citation: Théroux-Rancourt, G., M. R. Jenkins, C. R. Brodersen, A. McElrone, E. J. Forrestel, and J. M. Earles. 2020. Digitally deconstructing leaves in 3D using X-ray microcomputed tomography and machine learning. Applications in Plant Sciences 8(7): .   Description of the dataset A 'Cabernet Sauvignon' grapevine (Vitis vinifera L.) leaf from a plant of the BOKU experimental vineyard in Tulln, Austria, was scanned using microCT at the Swiss Light Source. The original reconstructions of the scans are using the gridrec (<a href="https://zenodo.org/api/files/bbca544a-15d0-40f3-8cc9-a3ee3c08fd7e/Gridrec_reconstruction_downsized.zip?versionId=28d98982-f69d-4eac-9dfa-efcc89c6823c">Gridrec_reconstruction_downsized.zip</a>) and the paganin, or phase-contrast, algortithm (<a href="https://zenodo.org/api/files/bbca544a-15d0-40f3-8cc9-a3ee3c08fd7e/Phase_contrast_reconstruction_downsized.zip?versionId=bef3260d-2865-4c9b-b5e1-e692edefb691">Phase_contrast_reconstruction_downsized.zip</a>). To facilitate automated segmentation, the size of the image in the x and y dimensions have been halved, so that the size of the pixels is 0.325 µm in those dimensions, but 0.1625 µm in the z (slices) dimension. A binary image segmenting the leaf cells and the airspace for each gridrec and phase-contrast stacks are created, and both are combined together (<a href="https://zenodo.org/api/files/bbca544a-15d0-40f3-8cc9-a3ee3c08fd7e/Binary_stack_for_local_thickness.zip?versionId=165e3938-b490-4e56-9c8e-a2084cb39d49">Binary_stack_for_local_thickness.zip</a>), a map of the local thickness is created (<a href="https://zenodo.org/api/files/bbca544a-15d0-40f3-8cc9-a3ee3c08fd7e/Local_thickness_map.zip?versionId=ce0a7dc7-5e3f-44a4-8881-cf84b6efd87c">Local_thickness_map.zip</a>). This map gives information on the largest diameter of the pixels labeled as cells in the binary stack. Hand-labeled slices or ground truths were drawn on the following slices: 80, 140, 200, 260, 340, 400, 440, 540, 620, 740, 800, 860, 940, 1060, 1140, 1240, 1300, 1400, 1480, 1540, 1600, 1690, 1740, 1840 (<a href="https://zenodo.org/api/files/bbca544a-15d0-40f3-8cc9-a3ee3c08fd7e/Hand_labelled_slices.tif?versionId=a21a13ac-fa47-4ef8-a903-ecc433787184">Hand_labelled_slices.tif</a>). Using the hand-labeled slices and the different images, a random-forest model was trained, which allowed to automatically segment the remaining slices of the stack (<a href="https://zenodo.org/api/files/bbca544a-15d0-40f3-8cc9-a3ee3c08fd7e/Fullstack_Prediction_Example-6_training_slices-6_testing_slices.zip?versionId=02b69e65-da85-492e-9b72-9b2b3ccd085f">Fullstack_Prediction_Example-6_training_slices-6...</a>). The source code for the segmentation program is available <a href="https://github.com/plant-microct-tools/leaf-traits-microct/tree/master">here</a>, and the source code for the testing used in the paper is available <a href="https://github.com/plant-microct-tools/leaf-traits-microct/tree/nb-slices-eval">here</a>.

opencc-by-4.0Oct 2020View details →

NEMO, HIDRA and Tide Gauge Datasets for HIDRA Machine Learning Algorithm Verification

Supporting sea level datasets for paper: "HIDRA 1.0: Deep-Learning-Based Ensemble Sea Level Forecastingin the Northern Adriatic" by Lojze Žust, Anja Fettich, Matej Kristan, and Matjaž Ličer

opencc-by-4.0Dec 2020View details →

FORCE 2020 Well well log and lithofacies dataset for machine learning competition

This well log dataset from 118 wells in the Norwegian Sea that has been used in the FORCE 2020 machine learning competition with seismic and wells to predict the lithofacies using machine learning models.  The well logs have been slightly cleaned up and partially despiked. The lithofacies and lithology interpretation has been hand crafted using skilled geoscientists  (Thanks to Explocrowd for excellent work). For citation in addition to the DOI please also refer to the github repository where the documentation and trained models reside <a href="https://github.com/bolgebrygg/Force-2020-Machine-Learning-competition">https://github.com/bolgebrygg/Force-2020-Machine-Learning-competition </a>   The original well log data comes form the Norwegian government and is provided by a NOLD 2.0 license

Machine learning reactive mixing dataset-0

Reactive-mixing dataset-0 for machine learning analyses

Machine learning reactive mixing dataset-2

Reactive-mixing dataset-1.1 for machine learning analyses

Machine learning reactive mixing dataset-1

Reactive-mixing dataset-1 for machine learning analyses

Machine learning reactive mixing dataset-9

Reactive-mixing dataset-3 for machine learning analyses

Machine learning reactive mixing dataset-4

Reactive-mixing dataset-1.3 for machine learning analyses

Machine learning reactive mixing dataset-3

Reactive-mixing dataset-1.2 for machine learning analyses