Find research datasets worth reusing

opencc-by-4.0Jun 2023View details →

data collection for machine learning using eddy force function data

Data for "On the choice of training data for machine learning of geostrophic mesoscale turbulence". Contains <ul> <li>qgm2 code (from James Maddison, University of Edinburgh) for generating the data</li> <li>codes to process data from qgm2 for machine learning, and train convolutional neural networks to use the data</li> <li>sample data</li> <li>processed data to reproduce plots in the paper</li> </ul> Files collected in different zip files to avoid the need to download the whole pack in one go. See readme for data/folder structure.

opencc-by-4.0Jun 2023View details →

Experiments data for "Reference Architecture for serverless Machine Learning"

Data collected during experiments in my bachelor thesis.

Data for the study: "Variational principle to regularize machine-learned density functionals: the non-interacting kinetic-energy functional"

This set of files contains the raw data generated by the study titled: "Variational principle to regularize machine-learned density functionals: the non-interacting kinetic-energy functional" Contains: - A set of Jupyter Notebooks to analyzed the data and produce the figures presented in the paper. - runs: Contains the training and validation scripts for each of the systems studied in this work. Also holds the model weights and validation data. Three folders are found: Hchain, noninteracting and Atoms. - datasets: Holds all the datasets generated and employed in this work.

Tailored machine learning models for functional RNA detection in genome-wide screens

The prediction of noncoding RNA and protein coding genetic loci has received   considerable attention in comparative genomics aiming in particular   at the identification of properties of nucleotide sequences that are   informative of their biological role in the cell. We present here a   software framework for the alignment-based training, evaluation and   application of machine learning models with user-defined   parameters. Instead of focusing on the one-size-fits-all approach of   pervasive \is annotation pipelines, we offer a framework for the   structured generation and evaluation of models based on arbitrary   features and input data, focusing on stable and explainable results.   Furthermore, we showcase the usage of our software package in a   full-genome screen of Drosophila melanogaster and evaluate   our results against the well-known but much less flexible program   RNAz.

Dataset for "Extraction of stratigraphic exposures on visible images using a supervised machine learning technique"

This is 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>Augmented images used in the U-Net training (aug_images.zip) <ul> <li>train/*.png: augmented original images (14,219 files)</li> <li>train_masks/*.png: augmented hand-masked images (14,219 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)

AneuploidyML_Datasets: Machine-learning analysis of factors that shape cancer aneuploidy landscapes reveals an important role for negative selection

This is raw data that were used to build the processed dataset for training and testing the machine learning

opencc-zeroAug 2023View details →

dryad32/100

Data for training AMSR2-CNN and its corresponding machine learning algorithm

Despite the availability of multiple decades of passive microwave measurements from satellite platforms, their utility for developing quantitative, spatially distributed estimates of snowpack is yet to be realized. A major bottleneck is the use of simple conceptual retrieval model formulations that are ineffective in representing the significant heterogeneity and complexity of snow evolution, particularly over areas with complex topography and forest regions. Here we demonstrate a physics-constrained and interpretable Convolutional Neural Network (CNN) to learn the functional relationship utilizing multi-channel passive microwave brightness temperature measurements from the Advanced Microwave Scanning Radiometer 2 (AMSR2) and in-situ snow depth observations. The machine learning approach with CNN generates vastly improved snow depth estimates relative to the standard AMSR2 estimates. Compared to independent in-situ measurements of snow depth over the Continental United States, the domain averaged Pearson correlation measure is three times higher than that of the standard AMSR2 estimates (R2: 0.68 versus 0.21), while the systematic errors are reduced by approximately fourfold. Further, the CNN-based snow depth estimates also exhibit notable enhancements in regions with forests, deep snow, and melting snow, thereby alleviating the limitations faced by traditional algorithms in retrieving accurate snow depths. The interpretation of the CNN framework further indicates that the machine learning approach dynamically leverages both volume scattering and emission components from a suite of measured passive microwave signals to generate more accurate snow depth retrievals. The results of this study provide an important benchmark of high-quality snow retrievals from passive microwave satellite measurements by maximizing their information content.

opencc-by-4.0May 2023View details →

Density Functional Theory and Machine Learning for Electrochemical Square-Scheme Prediction: An Application to Quinone-type Molecules Relevant to Redox Flow Batteries

The uploaded data contains (i) "01_Data" optimized molecular structure in XYZ format and the primary attributes and SMILES, (ii) "02_Datasets" datasets used in the publication, and (iv) "03_pynb_script" a Jupyter-Notebook. The 01_Data directory contains more than 8000 subdirectories. Each is for a molecule that undergoes a two-proton two-electron transfer reaction. In each subdirectory, one finds the following files: (1) directories named corresponding to the ones in Figure 1 of the paper. Inside each, there are geometries and properties in XYZ and CSV format, respectively. (2) "freeEnergy.dat" contains the free energy of different states. (3) "schemesquare.dat" has the parameters of the electrochemical scheme of square representation. ├── A │   ├── info.csv │   └── pos.xyz ├── A1- │   ├── info.csv │   └── pos.xyz ├── A2- │   ├── info.csv │   └── pos.xyz ├── AH │   ├── info.csv │   └── pos.xyz ├── AH1+ │   ├── info.csv │   └── pos.xyz ├── AH1- │   ├── info.csv │   └── pos.xyz ├── AH2 │   ├── info.csv │   └── pos.xyz ├── AH21+ │   ├── info.csv │   └── pos.xyz ├── AH22+ │   ├── info.csv │   └── pos.xyz ├── freeEnergy.dat └── schemesquare.dat ****************************************************** The new version (v1.1) contains some updates around: (i) The DFT calculations workflow in a folder called "04_workflow_of_DFT"           The Gaussian input files have been explained in the "README" file. (ii) The Python scripts for data extraction have been added and can be found in "05_how_to_extracted_data" (iii) We explained how to compute the Purbaix diagram in great detail "06_how_to_compute_Pourbaix_diagram/" All these changes/improvements were applied/made following the Referee of Digital Discovery Journal. Here, we would like to thank him/her.

opencc-by-4.0Aug 2023View details →

Machine learning-based detection of weather fronts with DL-FRONT in CESM1.3

These data are the results of detecting weather fronts with the machine learning algorithm DL-FRONT (see <a href="https://doi.org/10.5194/ascmo-5-147-2019">Biard and Kunkel 2019</a>) in simulations with the Community Earth System Model, version 1.3 (CESM1.3, see <a href="http://doi.org/10.1029/2019GL084057">Meehl et al. 2019</a>). The specific CESM1.3 simulations used here include: a historical climate simulation from 2000 to 2005, a simulation with Representative Concentration Pathway 2.6 (RCP2.6) forcing from 2006 to 2015, and a simulation with Representative Concentration Pathway 8.5 (RCP8.5) forcing from 2086–2100. See <a href="http://doi.org/10.1029/2022JD037038">Dagon et al. 2022</a> for a publication analyzing these machine learning detected weather fronts and associated extreme precipitation in historical and future climates. At each 3-hourly time step of simulation output over a North American spatial domain (10-77ºN, 171-31°W), DL-FRONT produced a set of spatial grids at 1° spatial resolution, for each of the five categories: cold front, warm front, stationary front, occluded front, and no front. Each cell in a spatial grid for a given category records the network-assigned probability (from 0.0 to 1.0) that the cell is in a weather front boundary region of that category (or, for the "no front" category, the probability that the cell is not in any weather front boundary region). The dataset contains three sets of files. The first set contains the original front probability maps. The second set contains "one hot" versions of the front probability maps. In the one hot version the five front-type probabilities for a spatial grid cell for a given time step are replaced by the value 1 for the largest front-type probability, and by 0 for the others. The third set contains front crossing rates (monthly, seasonal, annual) and climatologies/anomalies/standard deviations (monthly, seasonal) for each front type. The front probability files have names that follow the form cesm_fronts_<start_year>_<end_year>.nc for each simulation period. The one hot files have names that follow the form cesm_fronts__<start_year>_<end_year>_MaskedNetCDF_customgrid.nc.  The rates files have names that follow the form cesm_fronts_<start_year>_<end_year>_frontRates_viaPolylines_customgrid.nc. For the one-hot and rates files, the historical and RCP2.6 simulation output have been combined into a single file.

opencc-by-4.0Mar 2023View details →

Streamflow Predictions using Machine Learning with Data Reformation

Reference: Tran, Vinh Ngoc, Valeriy Y. Ivanov, and Jongho Kim. "Data reformation–A novel data processing technique enhancing machine learning applicability for predicting streamflow extremes." Advances in Water Resources 182 (2023): 104569. https://doi.org/10.1016/j.advwatres.2023.104569

opencc-by-4.0Sep 2023View details →

Dataset and Model Files for the Paper "Concrete Type Inference for Code Optimization using Machine Learning with SMT Solving"

Dataset and model files for the paper "Concrete Type Inference for Code Optimization using Machine Learning with SMT Solving".

opencc-by-4.0Nov 2022View details →

Dataset for manuscript "Turnover number predictions for kinetically uncharacterized enzymes using machine and deep learning"

Dataset for the github repository containing the code for the manuscript "Turnover number predictions for kinetically uncharacterized enzymes using machine and deep learning".

opencc-by-4.0Aug 2023View details →

Figures for 'Global streamflow modelling using process-informed machine learning'

High-quality figures for the article 'Global streamflow modelling using process-informed machine learning' (<a href="https://doi.org/10.2166/hydro.2023.217">https://doi.org/10.2166/hydro.2023.217</a>).

opencc-by-4.0Sep 2023View details →

data for "A machine-learned approach to monitor chemical reaction via in-situ infrared spectroscopy"

Source spectral and structural data of the AIMD trajectory and NEB calculation 1. <a href="https://zenodo.org/api/files/2d59762e-fb91-4c71-be73-fc9b18275144/180-structure-IR.zip">180-structure-IR</a>.zip Source spectral and structural data of the AIMD trajectory for the selected 180 configurations 2. <a href="https://zenodo.org/api/files/2d59762e-fb91-4c71-be73-fc9b18275144/md-pos-1.xyz?versionId=1a43760b-5cba-4548-85e6-6a45252403fc">md-pos-1.xyz</a> AIMD trajectories 3. <a href="https://zenodo.org/api/files/2d59762e-fb91-4c71-be73-fc9b18275144/ML-0-5100.tar.gz?versionId=ed65c63f-aa96-4be4-abef-e7a5bb7bf22f">ML-0-5100.tar.gz</a> <a href="https://zenodo.org/api/files/2d59762e-fb91-4c71-be73-fc9b18275144/ML-0-5100.tar.gz?versionId=ed65c63f-aa96-4be4-abef-e7a5bb7bf22f">ML-5101-7500.tar.gz</a> <a href="https://zenodo.org/api/files/2d59762e-fb91-4c71-be73-fc9b18275144/ML-0-5100.tar.gz?versionId=ed65c63f-aa96-4be4-abef-e7a5bb7bf22f">ML-7501-9500.tar.gz</a> Source spectral and structural data of the AIMD trajectory for the extracted 9500 configurations 4. <a href="https://zenodo.org/api/files/2d59762e-fb91-4c71-be73-fc9b18275144/split-neb-75.zip?versionId=a30f0f34-3ce2-4619-abd3-62e71174069b">split-neb-75.zip</a> Source spectral and structural data of the AIMD trajectory for the CO-CO  dimerization reaction

Quantifying the Effects of Vagus Nerve Stimulation on Gastric Myoelectric Activity in Ferrets Using an Interpretable Machine Learning Approach

Raw data

Machine Learning Estimation of Maximum Vertical Velocity from Radar (Test Data)

Inside this archive is the testing dataset for the paper titled: Machine Learning Estimation of Maximum Vertical Velocity from Radar, currently under review in AMS AIES. The files are compressed using tar.gz, so know that decompressed this data is about 20 GB in size. The training and validation sets are much larger. If you need those, please reach out to me and we will work on getting you the data. Please see the github repo for how to use this data: https://github.com/ai2es/hradar2updraft

Code for the article titled "A New Approach to Ozone Profiling Based on MAX-DOAS Observations and Stacking Machine Learning Models"

Open the record for dataset details and reuse information.

Machine learning based methods to generate conformational ensembles of disordered proteins (len54)

data is in rep_1 for all sequences, which contains the trajectory (xtc) file, Rg information (in the file Rg.out), pairwise distance information (in the file traj_analysis_data/pairwise_distance_matrix.csv) and the bspline coefficients (in the file bspline_info/xyz_coeff.npy)

Machine learning based methods to generate conformational ensembles of disordered proteins (len18, point mutation)

data is organized by bin number (0-9) and mutation location (4, 8, 12). (Note: in the manuscript, we used the nomenclature bins 1-10 and mutation locations 5,8,13. We simply used a 0-index convention when naming our folders). All data is in rep_1, which contains the trajectory (xtc) file, Rg information (in the file Rg.out), pairwise distance information (in the file traj_analysis_data/pairwise_distance_matrix.csv) and the bspline coefficients (in the file bspline_info/xyz_coeff.npy). However, for bin_num=9/mutation_loc=12, the data used is in rep_2, not rep_1.

Machine learning based methods to generate conformational ensembles of disordered proteins (len36)