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.
921
datasets available to search
ShareScore release 0.7.1
Dataset results
921 results for “neural networks”
Data of A recurrent neural network-accelerated multi-scale model for elasto-plastic heterogeneous materials subjected to random cyclic and non-proportional loading paths
<pre>Data related to the publication (we would be grateful if you could cite the paper in the case in which you are using the data) title = "A recurrent neural network-accelerated multi-scale model for elasto-plastic heterogeneous materials subjected to random cyclic and non-proportional loading paths", journal = "Computer Methods in Applied Mechanics and Engineering", pages = " 113234", year = "2020", issn = "0045-7825", doi = "https://doi.org/10.1016/j.cma.2020.113234", author = "Wu, Ling and Nguyen, Van Dung and Kilingar, Nanda Gopala and Noels, Ludovic"</pre>
CellCognize: a neural network pipeline for cell type classification from flow cytometry data
<p>Readme file content</p> <p>The files stored here contain the following material as supplementary and source data for the publication</p> <p>Rapid detection of microbiota cell type diversity using machine-learned classification of flow cytometry data</p> <p>Birge D. Özel Duygan1, Noushin Hadadi1, Ambrin Farizah Bab1, Markus Seyfried2, Jan R. van der Meer1</p> <p>1 Department of Fundamental Microbiology, University of Lausanne, 1015 Lausanne, Switzerland<br> 2 Biotechnology Department, Firmenich SA, Geneva, Switzerland</p> <p>%%%%%%%<br> Flow cytometry data<br> %%%%%%</p> <p>FCM_files:</p> <p>.mat files with cleaned data as described in the supplementary methods section</p> <p>Ecoli_lakewater: raw FCM data (in .csv format) of E. coli cultures and E. coli cultures mixed to lakewater</p> <p>MIX_experiment_ACL_AJH_PVR: raw FCM data (in .csv format) of the synthetic three culture experiment with E. coli, A. johnsonii and P. veronii, as described in the main text and SI methods.</p> <p>PHE_OCT_enrichments: raw FCM data (in .csv format) of the phenol and 1-octanol enrichments and the 1-octanol isolates, as described in the main text and SI methods.</p> <p>%%%%%%%<br> Neural network data<br> %%%%%%</p> <p>NN_file_example: three ANN functions, to be used in conjunction with the SI methods section</p> <p>Supplementary_Methods.docx: Detailed description on the construction, usage and scripts for the ANN. To be used in conjunction with the Flow Cytometry data</p> <p>%%%%%%%<br> 16S sequencing data<br> %%%%%%</p> <p>raw fastq- files of the sample reads of the 1-octanol and phenol enrichments described in the paper, at t=0 and t=3d, each in triplicates, forward and reverse.</p> <p>Readme_16S_sequence_files.txt: sample description of the read files</p>
Supplementary Materials (An Artificial Neural Network Model for Assessing Frailty-Associated Factors in the Thai Population)
<p>Supporting information for an Artificial Neural Network Model for Assessing Frailty-Associated Factors in the Thai Population</p>
Classify on the Clock (CloCk) - An Entry Level Image Data Set for Neural Networks
<p><strong>General information</strong></p> <p>This data set contains synthetic images of an analog clock. Each time point from 00:00:00 - 11:59:59 is included as a separate image. We provide three different versions of each image with an increasing number of additional information:</p> <ul> <li><em>Sparse</em>: The image includes just the hour, minute and second hand</li> <li><em>Reduced</em>: Additional markings around the clock</li> <li><em>Full</em>: Additional written numbers</li> </ul> <p>The hands differ in size and color:</p> <ul> <li><em>Hour</em>: Black, short, wide</li> <li><em>Minute</em>: Blue, long, medium</li> <li><em>Second</em>: Red, long, slim</li> </ul> <p>We provide the following files in this data repository:</p> <ul> <li>RGB images with a size of 512x512 for all three versions</li> <li>A suggested training, validation and test split</li> <li>The creation script as Python file</li> </ul> <p>The script can easily be modified to create images with a different size and color.</p> <p> </p> <p><strong>Clock System</strong></p> <p>The movements of the hands follow a linear relationship. Their angles can be calculated by the forward system:</p> <p><span class="math-tex">\( \begin{bmatrix} 6^\circ & 0 & 0 \\ 0.1^\circ & 6^\circ & 0 \\ 0 & 0.5^\circ & 30^\circ \end{bmatrix} \begin{pmatrix} n_{\text{sec}} \\ n_{\text{min}} \\ n_{\text{hour}} \end{pmatrix} = \begin{pmatrix} \alpha_{\text{sec}}\\ \alpha_{\text{min}} \\ \alpha_{\text{hour}} \end{pmatrix}.\)</span></p> <p>One can also introduce rotated versions of the images. The system becomes non-linear in this case:</p> <p><span class="math-tex">\(\operatorname{mod}\left( \begin{bmatrix} 6^\circ & 0 & 0 \\ 0.1^\circ & 6^\circ & 0 \\ 0 & 0.5^\circ & 30^\circ \end{bmatrix} \begin{pmatrix} n_{\text{sec}} \\ n_{\text{min}} \\ n_{\text{hour}} \end{pmatrix} + \omega, \, 360^\circ \right) = \begin{pmatrix} \alpha_{\text{sec}}\\ \alpha_{\text{min}} \\ \alpha_{\text{hour}} \end{pmatrix}.\)</span></p> <p> </p> <p><strong>Use Cases</strong></p> <p>This data set was originally designed for basic research on Capsule Networks. Use cases are:</p> <ul> <li>Evalutation of classification and regression performance</li> <li>Influence of image transformations, e.g. rotations</li> <li>Detection of the hierarchy & relationship of image parts</li> <li>Concealment of objects in the image</li> <li>Interpretability of the learned model</li> <li>Solving a discrete inverse problem (<em>sparse</em> version)</li> <li>...</li> </ul>
Predicting Phenotype from Multi-Scale Genomic and Environment Data using Neural Networks and Knowledge Graphs
<p><strong>Background: To mitigate the effects of climate change on public health and conservation, we need to better understand the dynamic interplay between biological processes and environmental effects. Machine learning (ML) methods in general, and Deep Learning (DL) methods in particular, are a potential way forward because they are able to cope with the nonlinearity of natural systems. However, there are several barriers that exist, including the absence of ML-ready data. We propose to develop a machine learning framework capable of predicting phenotypes based on multi-scale data about genes and environments. A critical part of this framework are data transformation methods that map the heterogeneous input data into formats that are consumable by the ML techniques. The central hypothesis of this research is that deep learning algorithms and biological knowledge graphs will predict phenotypes more accurately across more taxa and more ecosystems than do current numerical and traditional statistical modeling methods. Our long term goal is to develop predictive analytics for organismal response to environmental perturbations using innovative data science approaches. This pilot project on predicting emergent properties of complex systems and multidimensional interactions is funded by the NSF (Award # 1939945, 1940059, 1940062, 1940330). </strong></p> <p> </p> <p><strong>Results: We have established shared project governance, communication channels, project timeline, and data and computing environment across four universities. We have successfully reached out to three other projects for broader collaboration.</strong></p>
Data from: Chromosome-scale inference of hybrid speciation and admixture with convolutional neural networks
<p>Inferring the frequency and mode of hybridization among closely related organisms is an important step for understanding the process of speciation and can help to uncover reticulated patterns of phylogeny more generally. Phylogenomic methods to test for the presence of hybridization come in many varieties and typically operate by leveraging expected patterns of genealogical discordance in the absence of hybridization. An important assumption made by these tests is that the data (genes or SNPs) are independent given the species tree. However, when the data are closely linked, it is especially important to consider their non-independence. Recently, deep learning techniques such as convolutional neural networks (CNNs) have been used to perform population genetic inferences with linked SNPs coded as binary images. Here we use CNNs for selecting among candidate hybridization scenarios using the tree topology (((P<sub>1</sub>,P<sub>2</sub>),P<sub>3</sub>),Out) and a matrix of pairwise nucleotide divergence (d<sub>XY</sub>) calculated in windows across the genome. Using coalescent simulations to train and independently test a neural network showed that our method, HyDe-CNN, was able to accurately perform model selection for hybridization scenarios across a wide-breath of parameter space. We then used HyDe-CNN to test models of admixture in <em>Heliconius</em> butterflies, as well as comparing it to a random forest classifier trained on introgression-based statistics. Given the flexibility of our approach, the dropping cost of long-read sequencing, and the continued improvement of CNN architectures, we anticipate that inferences of hybridization using deep learning methods like ours will help researchers to better understand patterns of admixture in their study organisms.</p>
MOLI: multi-omics late integration with deep neural networks for drug response prediction
<p>Harmonized data used in "MOLI: multi-omics late integration with deep neural networks for drug response prediction", 2019, <em>Bioinformatics </em><a href="https://academic.oup.com/bioinformatics/article/35/14/i501/5529255">https://academic.oup.com/bioinformatics/article/35/14/i501/5529255</a>. <br> CNA.tar.gz contains CNA profiles with non-integer estimates of copy number, e.g. log-ratios. Please use binarized CNA profiles (CNA_binary.tar.gz) to replicate the results described in the paper. </p> <p><br> All raw data were obtained from open sources:<br> - https://www.cancerrxgene.org/<br> - ArrayExpress https://www.ebi.ac.uk/arrayexpress/<br> - Firehose Broad GDAC http://gdac.broadinstitute.org/runs/stddata__2016_01_28/data/<br> - Supplementary of Gao et al., 2015 https://www.nature.com/articles/nm.3954</p> <p>Gene symbols were mapped to Entrez Gene IDs. Data preprocessing is described in detail in supplementary materials. The code is available at <a href="https://github.com/hosseinshn/MOLI/tree/master/preprocessing_scr">https://github.com/hosseinshn/MOLI/tree/master/preprocessing_scr</a>.</p>
Approximation of a marine ecosystem model by artificial neural networks designed using a genetic algorithm
<p>Data from the Paper: Approximation of a marine ecosystem model by artificial neural networks designed using a genetic algorithm.</p> <p>Abstract: </p> <p>Marine ecosystem models are important to identify the processes that affects for example the global carbon cycle. Computation of an annually periodic solution (i.e., a steady annual cycle) for these models requires a high computational effort. To reduce this effort, we approximated an exemplary marine ecosystem model by different artificial neural networks. We used a fully connected network, then applied the sparse evolutionary training (SET) procedure, and finally applied a genetic algorithm (GA) to optimize both the network topology. With all three approaches, a direct approximation of the steady annual cycle was not sufficiently accurate. However, using the mass-corrected prediction of the ANN as initial concentration for additional model runs, the results were in very good agreement. In this way, we achieved a runtime reduction by about 15 \%. The result from the SET algorithm were comparable to those of the full network. Further application of the GA may lead to an even higher reduction.</p> <p>Content:</p> <p>Database sqlite <a href="https://zenodo.org/api/files/669d208b-7304-4d31-a6cc-3ad78e3544e9/ANN_Database.db">ANN_Database.db</a></p> <p>zip-files with data: </p> <p><a href="https://zenodo.org/api/files/669d208b-7304-4d31-a6cc-3ad78e3544e9/ANN-Data.zip">ANN-Data.zip</a> structure and weights of used networks</p> <p><a href="https://zenodo.org/api/files/669d208b-7304-4d31-a6cc-3ad78e3544e9/ANN-Results.zip">ANN-Results.zip</a> results obtained with networks</p> <p><a href="https://zenodo.org/api/files/669d208b-7304-4d31-a6cc-3ad78e3544e9/Reference-Results.zip">Reference-Results.zip</a> reference results and training data</p> <p> </p> <p> </p> <p> </p> <p> </p>
Recurrent Neural Network Language Models Always Learn English-Like Relative Clause Attachment
<p>This repository contains the raw results (by word information-theoretic measures for the experimental stimuli) and the LSTM models analyzed in <a href="https://www.aclweb.org/anthology/2020.acl-main.179/">Recurrent Neural Network Language Models Always Learn English-Like Relative Clause Attachment</a>. The models from the synthetic experiments are given in the synthetic archive, as well as the training data generation script. There is a README included that gives more details for recreating/evaluating results from those experiments.</p> <p>The naming convention for each model in the models directory is:<br> [Language]_hidden[Hidden Units]_batch[Batch Size]_dropout[Dropout Rate]_lr[Learning Rate]_[Model Number].pt</p> <p>Language: en for English and es for Spanish<br> Hidden Units: All models had two layers with 650 hidden units per layer<br> Batch Size: The size of the batch (128 for English, 64 for Spanish)<br> Dropout Rate: All models used a dropout rate of 0.2<br> Learning Rate: All models has a learning rate of 20<br> Model Number: Identifier of the model (English model 0 is the best model from <a href="https://github.com/facebookresearch/colorlessgreenRNNs">Gulordava et al. (2018)</a>) </p> <p> </p> <p> </p>
Data/code for Interpretable, calibrated neural networks for analysis and understanding of neutron spectra
<p>Simulated neutron spectra from inelastic neutron scattering on the double perovskite PCSMO. The dataset is related to the publication <em>Interpretable, calibrated neural networks for analysis and understanding of neutron spectra</em>. The paper is yet to be submitted, but the references will be added in due course.</p> <p># Pre-trained model weights</p> <p>`model-weights.tgz`</p> <p>This file contains the weights for the models reported in the paper which can be downloaded and used to re-produce the results.</p> <p>Please download the `model-weights.tgz` file and extract it to the root of the repository `interpretable-ml-neutron-spectroscopy` from github.</p> <p># Data Generation</p> <p>`data_generation.tar.gz`</p> <p> Large data files for github repository associated with the publication _Interpretable, calibrated neural networks for analysis and understanding of neutron spectra_ These files relate to generation of the data in the paper.</p> <p>Please download the `data_generation.tar.gz` file and extract it to the root of the repository `interpretable-ml-neutron-spectroscopy` from github.</p> <p>#Training data</p> <p>`<a href="https://zenodo.org/api/files/47a40b89-ae6f-4655-aa4b-2d8fc11ec597/pcsmo-simulated-spectra.tgz?versionId=1ff107a9-85b9-4b46-9c77-84319ca9d59a">pcsmo-simulated-spectra.tgz</a>`</p> <p>Contains the data used to train the networks. Extract this file to a data directory and then point the `train.py` files for the various models to look for this `<datadir>` in the specified location in those files.</p>
Data for assessment of damage to residential dwellings using artificial neural networks
<p>The data provided and the associated MATLAB code were used to build an Artificial Neural Network Model to capture damage to residential home subjected to tornado events in the State of Missouri. The ANN model utilizes relevant tornado, societal demographic, and structural data to determine a building's resulting damage state from an extreme wind event. </p>
Greenland ice sheet surface melt projections using artifical neural networks
<p>Surface melt projections for the Greenland ice sheet using artificial neural networks.</p> <p>Organized as follows:</p> <p><scenario>/<variable>/<model>/<ensemble_number>.nc</p>
Data for "CryoDRGN: Reconstruction of heterogeneous cryo-EM structures using neural networks"
<p>Trained models and reconstructed density maps for:</p> <ul> <li>EMPIAR-10028: "Cryo-EM structure of a <em>Plasmodium falciparum</em> 80S ribosome bound to the anti-protozoan drug emetine" from Wong et al. (2014)</li> <li>EMPIAR-10049: "Molecular Mechanism of V(D)J Recombination from Synaptic RAG1-RAG2 Complex Structures" from Ru et al. (2015)</li> <li>EMPIAR-10076: "Modular assembly of the large bacterial ribosome" from Davis et al. (2016)</li> <li>EMPIAR-10180: "Structure of a pre-catalytic spliceosome" from Plaschka et al. (2017)</li> </ul> <p>Synthetic datasets with simulated heterogeneity and their ground truth density maps, poses, and labels:</p> <ul> <li>Uniform: 50k particle images (128x128, 6A/pix) uniformly sampled from 50 models along a 1-dimensional reaction coordinate</li> <li>Cooperative: 50k particle images (128x128, 6A/pix) sampled along the above reaction coordinate according to a 3-component Gaussian mixture model with overlapping components </li> <li>Noncontiguous: 50k particle images (128x128, 6A/pix) sampled along the above reaction coordinate according to a 3-component Gaussian mixture model without overlapping components</li> <li>Ribosomes: 50k particle images (128x128, 3A/pix) containing a mixture of 30S, 50S, 70S ribosomes simulating compositional heterogeneity</li> </ul>
Supplementary Material: Exposing Previously Undetectable Faults in Deep Neural Networks
<p>Supplementary material for ISSTA 2021 submission #58. The supplementary material is under 600MB and uploaded before the submission deadline, but uploading to the submission website did not work. Editing this supplementary material is not possible after upload.</p>
Double attention recurrent convolution neural network for answer selection
<p>Answer selection is one of the key steps in many Question Answering (QA) applications. In this paper, a new deep model with two kinds of attention is proposed for answer selection: the Double Attention Recurrent Convolution Neural Network (DARCNN). Double attention means self-attention and cross-attention. The design inspiration of this model came from the Transformer in the domain of machine translation. Self-attention can directly calculate dependencies between words regardless of the distance. However, self-attention ignores the distinction between its surrounding words and other words. Thus, we design a decay self-attention that prioritizes local words in a sentence. In addition, cross-attention is established to achieve interaction between question and candidate answer. With the outputs of self-attention and decay self-attention, we can get two kinds of interactive information via cross-attention. Finally, using the feature vectors of the question and answer, elementwise multiplication is used to combine with them and multi-layer perceptron (MLP) is used to predict the matching score. Experimental results on four QA datasets containing Chinese and English show that DARCNN performs better than other answer selection models, thereby demonstrating the effectiveness of self-attention, decay self-attention and cross-attention in answer-selection tasks.</p>
Neural network weights for the xrv.baseline_models.chexpert.DenseNet model for torchxrayvision
<p>https://github.com/mlmed/torchxrayvision</p> <p>xrv.baseline_models.chexpert.DenseNet</p>
Code and data for "An integrated microwave neural network for broadband computation and communication"
<div> <div> </div> </div> <div> <div> <div> <div> <div> <div> <p>This repository contains code and data used in the presentation of results in the article "An integrated microwave neural network for broadband computation and communication". The contents of the zipped files are:</p> <ul> <li><strong>Spectrum Analyzer Outputs (Datasets and ML scripts for digital emulation, radar and signal encoding classification.zip)</strong>: Reduced-bandwidth outputs used to train the backend for results presented in Figs. 3 and 4 and Supplementary Fig. 3.</li> <li><strong>Simulation Code (Coupled mode simulation of integrated MNN.zip) </strong>: For modeling the coupled MNN system shown in Fig. 2 and Extended Figs. 4 and 5.</li> <li><strong>Radar Signal Simulation (Training data and code for simulating dynamic targets in simulated airspace.zip)</strong>: Code to simulate received baseband signals from radar targets.</li> </ul> <p>Each folder contains readme files on how to run the code and analyze data.</p> <p>Please install a recent Python release (https://www.python.org/downloads/) and a recent release of MATLAB (https://www.mathworks.com/help/install/) to run the code. No non-standard hardware is required. </p> </div> </div> </div> </div> </div> </div>
Input Data for A Fast Surrogate Model for 3D-Earth Glacial Isostatic Adjustment using Tensorflow (v2.8.0) Artificial Neural Networks
<p>Training datasets for the manuscript A Fast Surrogate Model for 3D-Earth Glacial Isostatic Adjustment using Tensorflow (v2.8.0) Artificial Neural Networks. Two separate datasets are contained for training the ANNs: the 3D-spherically-symmetric (SS) rate-of-change of relative sea level (ROCRSL) and the 3D-SS rate of change of radial displacement (ROCRAD) as a function of SS profiles. Two other datasets contain RSL projections from the explicit (i.e. Seakon 3D - Seakon SS + NMSS ) model and the NMSS model, labelled Seakon_plus_NMSS_RSL and NMSS respectively.</p> <p>Filenames denote the structure of the SS profile: </p> <p>???_?.??_??.*.csv = LT_UMV_LMV.*.{csv,nc}<br> </p> <p>LT = elastic lithosphere thickness (km)</p> <p>UMV = upper mantle viscosity (1E21 Pa s)</p> <p>LMV = lower mantle viscosity (1E21 Pa s)</p> <p>i.e. 96_0.5_10.seakon_S40RTS_lr18-SS.rrad.roc.r360x180.P5.density_wSSRRADROC.csv.bz2 has the SS profile</p> <p>96km elastic lithosphere, 0.5E21 Pa s upper mantle viscosity, 10E21 Pa s lower mantle viscosity</p> <p> </p> <p>The columns of the input files are as follows:</p> <p>LT, UMV, LMV, longitude, latitude, time(t=0), ice(t=0), SS_ROC_RSL (t=0), time(t=-1), ice(t=-1), time(t=-2), ice(t=-2), time(t=-3), ice(t=-3), time(t=-4), ice(t=-4), 3D-SS_ROC_RSL(t=0)</p> <p>units for the above are as follows:</p> <p>km, 1E21 Pas, 1E2 Pas, degrees east (0->360), degrees (-180->180), days since 2000, m, mm/year, days since 2000, m, days since 2000, m, days since 2000, m, days since 2000, m, mm/year</p> <p>where 'days since 2000' assumes exactly 365.25 days per year.</p>
BigVSAN: Enhancing GAN-based Neural Vocoders with Slicing Adversarial Network
<p>This repository contains a pre-trained checkpoint for BigVSAN proposed in the paper <a href="https://arxiv.org/abs/2309.02836">BigVSAN: Enhancing GAN-based Neural Vocoders with Slicing Adversarial Network</a> by Sony.</p><p>More information about BigVSAN including our code is available at <a href="https://github.com/sony/bigvsan">https://github.com/sony/bigvsan</a>.</p>
Emulating present and future simulations of melt rates at the base of Antarctic ice shelves with neural networks
<p>This dataset contains the data and scripts for the publication "<a href="https://doi.org/10.1029/2023MS003829">Emulating present and future simulations of melt rates at the base of Antarctic ice shelves with neural networks</a>" in <em>Journal of Advances in Modeling Earth Systems</em>.</p> <p>Before going into details, here is a reminder that the NEMO runs for the training dataset are called 'OPM+number'. These are the corresponding names given in the manuscript: OPM006=HIGHGETZ, OPM016=WARMROSS, OPM018=COLDAMU and OPM021=REALISTIC. For the testing dataset: 'bf663' is the REPEAT1970 run and 'bi646' is the 4xCO2 run.</p> <p>Most of the formatting and preprocessing of the training data has been made for <a href="https://tc.copernicus.org/articles/16/4931/2022/">Burgard et al. 2022</a>. The raw and to-some-degree processed data can therefore be found here: <a href="https://doi.org/10.5281/zenodo.7308352">https://doi.org/10.5281/zenodo.7308352</a>. <br>The raw data for the testing dataset is from <a href="https://doi.org/10.1029/2021MS002520">Smith et al. 2021</a>, you can find it here: <a href="doi.org/10.5281/zenodo.7886986">https://doi.org/10.5281/zenodo.7886986</a></p> <p>The following folders and files can be found here:</p> <p>===============<br><strong>raw/</strong></p> <p>Some geometrical files needed for initial data formatting and masking.</p> <p>===============<br><strong>interim/</strong></p> <ul> <li><strong>ANTARCTICA_IS_MASKS</strong>/ (<em>from INTERIM_ANTARCTICA_IS_MASKS.zip</em>): contains <ul> <li>masks and geometric information for the testing dataset to be included in the input file of the neural network and for the classic parameterisations.</li> <li>local bedrock and ice meridional and zonal slopes</li> </ul> </li> <li><strong>BOXES/</strong> (<em>from INTERIM_BOXES.zip</em>): contains variables needed to apply the box parameterisation for the testing dataset</li> <li><strong>PLUMES/ </strong>(<em>from INTERIM_PLUMES.zip</em>): contains the variables needed to apply the plume parameterisation for the testing dataset</li> <li><strong>SMITH_bf663/</strong><em><strong> and </strong></em><strong>SMITH_bi646/ </strong>(<em>from INTERIM_SMITH*.zip</em>): for testing dataset, <ul> <li>corrected_draft_bathy_isf.nc: file containing ice draft and bathymetry corrected by ice draft concentration to account for the biased draft and bathymetry at the grounding line resulting from the interpolation from the native NEMO grid to the stereographic grid (values under ice shelf and NaNs over land</li> <li>custom_lsmask_Ant_stereo_clean.nc: land-sea mask giving 0 = ocean, 1 = shelf, 2 = land</li> <li>isfdraft_conc_Ant_stereo.nc: ice-shelf concentration resulting from the interpolation from the native NEMO grid to the stereographic grid</li> <li>other_mask_vars_Ant_stereo.nc: contains other variables used for the masks</li> <li>the reference melt: 1D containing integrated melt, 2D containing melt fields, box1 containing melt near the grounding line</li> </ul> </li> <li><strong>T_S_PROF/ </strong>(<em>from INTERIM_T_S_PROF.zip</em>) <ul> <li>Mean profiles used as input for traditional parameterisations</li> <li>T and S 2D fields, extrapolated from the mean profiles to the local ice draft depth (needed as input for the neural network)</li> <li>Fields of mean and standard deviation T and S for all points (needed as input for the neural network)</li> </ul> </li> <li><strong>NN_MODELS/</strong><em><strong> </strong>(from </em>INTERIM_NN_MODELS<em>.zip</em>) contains all neural networks trained for this paper (for the cross validation and over the whole dataset for testing)</li> <li><strong>INPUT_DATA/ </strong>(from INTERIM<em>_</em>INPUT_DATA.zip) contains all input csv files containing the input datasets for the different training and testing iterations. Also contains the metrics to normalise the input. For the cross-validation, the input csv files are not included because they are too large. However, they can be reconstructed from the individual files for ice shelves and time blocks. The metrics to normalise the data during the cross-validation are included in EXTRAPOLATED_ISFDRAFT_CHUNKS_CV!</li> </ul> <p>===============<br><strong>processed/MELT_RATE/</strong></p> <p>Contains resulting melt rates</p> <ul> <li><strong>CV_ISF :</strong> Cross-validation results over ice shelves</li> <li><strong>CV_TBLOCKS : </strong>Cross-validation results over time</li> <li><strong>SMITH_bf663 : </strong>Neural network results for REPEAT1970</li> <li><strong>SMITH_bf663_CLASSIC : </strong>"Traditional" parameterisation results for REPEAT1970</li> <li><strong>SMITH_bi646 :</strong> Neural network results for 4xCO2</li> <li><strong>SMITH_bi646_CLASSIC:</strong> "Traditional" parameterisation results for 4xCO2</li> </ul> <p>=====================</p> <p>The explanation around the scripts can be found in README.rst with the scripts in<strong> scripts_paper_simpleNN_basal_melt.zip</strong>.<br><em>Note that these are the scripts needed to produce the results in the paper. You can also find them on Github: </em><a href="https://github.com/ClimateClara/https://github.com/ClimateClara/scripts_paper_simpleNN_basal_melt"><em>https://github.com/ClimateClara/scripts_paper_simpleNN_basal_melt</em></a>, <em>find the most up-to-date version of the package 'multimelt' here: </em><a href="https://github.com/ClimateClara/multimelt"><em>https://github.com/ClimateClara/multimelt</em></a><em> and a version you can install via pip here: </em><a href="https://github.com/ClimateClara/multimelt"><em>https://pypi.org/project/multimelt/</em></a></p> <p>Finally, if anything is unclear, check out the "Methods" section of the paper: <a href="https://doi.org/10.1029/2023MS003829">https://doi.org/10.1029/2023MS003829</a></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.