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.
25
datasets available to search
ShareScore release 0.9.0
Dataset results
25 results for “Network Representation”
Self-Supervised Representation Learning on Neural Network Weights for Model Characteristic Prediction - Datasets
<p>Datasets to NeurIPS 2021 accepted paper "Self-Supervised Representation Learning on Neural Network Weights for Model Characteristic Prediction".</p> <p>Datasets are pytorch files containing a dictionary with training, validation and test sets. Train, validation and test sets are custom dataset classes which inherit from the standard torch dataset class. Corresponding code an be found at https://github.com/HSG-AIML/NeurIPS_2021-Weight_Space_Learning.</p> <p>Datasets 41, 42, 43 and 44 are our dataset format wrapped around the zoos from Unterthiner et al, 2020 (https://github.com/google-research/google-research/tree/master/dnn_predict_accuracy)<br> <br> Abstract:<br> Self-Supervised Learning (SSL) has been shown to learn useful and information-preserving representations. Neural Networks (NNs) are widely applied, yet their weight space is still not fully understood. Therefore, we propose to use SSL to learn neural representations of the weights of populations of NNs. To that end, we introduce domain specific data augmentations and an adapted attention architecture. Our empirical evaluation demonstrates that self-supervised representation learning in this domain is able to recover diverse NN model characteristics. Further, we show that the proposed learned representations outperform prior work for predicting hyper-parameters, test accuracy, and generalization gap as well as transfer to out-of-distribution settings.</p>
CNN Wild Park - Graph Neural Networks for Learning Equivariant Representations of Neural Networks
<p>This repository contains the <strong>CNN Wild Park</strong> dataset from the paper:</p> <blockquote> <p><strong>Graph Neural Networks for Learning Equivariant Representations of Neural Networks</strong><br><a href="https://mkofinas.github.io/">Miltiadis Kofinas</a>*, <a href="https://bknyaz.github.io/">Boris Knyazev</a>, <a href="https://www.cyanogenoid.com/">Yan Zhang</a>, <a href="https://yunlu-chen.github.io/">Yunlu Chen</a>, <a href="https://gertjanburghouts.github.io/">Gertjan J. Burghouts</a>, <a href="https://egavves.com/">Efstratios Gavves</a>, <a href="https://www.ceessnoek.info/">Cees G. M. Snoek</a>, <a href="https://davzha.netlify.app/">David W. Zhang</a>*<br><em>ICLR 2024</em> (oral)<br><a href="https://arxiv.org/abs/2403.12143">https://arxiv.org/abs/2403.12143</a><br><a href="https://github.com/mkofinas/neural-graphs">https://github.com/mkofinas/neural-graphs</a><br>*Joint first and last authors</p> </blockquote> <p>We introduce a new dataset of CNNs, which we term <em>CNN Wild Park</em>.<br>The dataset consists of 117,241 checkpoints from 2,800 CNNs, trained for up to 1,000 epochs on CIFAR10.<br>The CNNs vary in the number of layers, kernel sizes, activation functions, and residual connections between arbitrary layers.</p> <p>More specifically, we construct the CNN Wild Park dataset by training 2,800 small CNNs with different architectures for 200 to 1,000 epochs on CIFAR10. We retain a checkpoint of its parameters every 10 steps and also record the test accuracy. The CNNs vary by:</p> <ul> <li>Number of layers L in [2, 3, 4, 5] (note that this does not count the input layer).</li> <li>Number of channels per layer c_l in [4, 8, 16, 32].</li> <li>Kernel size of each convolution k_l in [3, 5, 7].</li> <li>Activation functions at each layer are one of ReLU, GeLU, tanh, sigmoid, leaky ReLU, or the identity function.</li> <li>Skip connections between two layers with at least one layer in between. Each layer can have at most one incoming skip connection. We allow for skip connections even in the case when the number of channels differ, to increase the variety of architectures and ensure independence between different architectural choices. We enable this by adding the skip connection only to the min(c_n, c_m) nodes.</li> </ul> <p>We divide the dataset into train/val/test splits such that checkpoints from the same run are <strong>not</strong> contained in both the train and test splits. </p> <div> </div> <div> </div>
Joint representation of molecular networks from multiple species improves gene classification - Data
<p>This is the data the accompanies the manuscript <em>Joint representation of molecular networks from multiple species improves gene classification</em></p> <p>Below is the license agreement for each of the publicly available datasets </p> <ul> <li><a href="https://wiki.thebiogrid.org/doku.php/terms_and_conditions">BioGRID</a></li> <li><a href="http://geneontology.org/docs/go-citation-policy/">GO</a></li> <li><a href="https://www.disgenet.org/legal">DisGeNet</a></li> <li><a href="https://monarchinitiative.org/about/licensing">Monarch</a></li> <li><a href="http://eggnog-mapper.embl.de">eggNOG</a></li> </ul> <p>No license agreement was available on <a href="http://imp.princeton.edu">IMP web site</a>, however we have obtained permission from the owner of the material to redistribute the network.</p>
PGB: A PubMed Graph Benchmark for Heterogeneous Network Representation Learning
<p>PubMed Graph Benchmark (PGB) aggregates the metadata associated with the biomedical articles from PubMed into a unified source. The benchmark contains metadata including title, abstract, authors, in/out citations, MeSH terms, MeSH hierarchy, venue, publication type, and chemicals.</p>
Comparing the representation of a simple visual stimulus across the cerebellar network [dataset]
<p>This repository contains all the raw data that has been used to generate the figures in the paper "Comparing the representation of a simple visual stimulus across the cerebellar network".<br> To use the dataset, select all 10 .zip files provided here and unpack them simultaneously.</p> <p> </p> <p>The dataset contains the following folders:</p> <ul> <li>freely_swimming_beh folder contains the data for the freely swimming behavior experiments. Inside, subfolders for each behavioral experiment run can be found, containing the experiment metadata and the behavioral log.</li> <li>flashes folder contains the data for the imaging experiments with the FLASHES protocol. Within this folder, three subfolders can be found with the data for each one of the three different types of imaged cell types (GCs, IONs and PCs). Within those subfolders, a folder for each imaged fish can be found, containing a data dictionary with the extracted ROI traces, and a folder with the experiment metadata.</li> <li>steps folder contains the data for the imaging experiments with the STEPS protocol. Within this folder, three subfolders can be found with the data for each one of the three different types of imaged cell types (GCs, IONs and PCs). Within those subfolders, a folder for each imaged fish can be found, containing a data dictionary with the extracted ROI traces, and a folder with the experiment metadata.</li> <li>fitting folder contains data dictionaries with the results from the model fitting needed to reproduce the manuscript figures.</li> <li>data_dict4decoding_complete.h5 is a dictionary file containing all necessary data to reproduce the decoding experiments.</li> </ul>
Figure 3. Neural network representation for HMM given in Figure 2.-Neuroevolution Mechanism for Hidden Markov Model
<p>In our proposed structure, we injected a hidden layer to have a multilayer perceptron which<br> is more efficient than single layer perceptron.<br> To make this process clear, Figure 3 shows the neural networks for the HMM presented in<br> Figure 2.</p>
Local synaptic inputs support opposing, network-specific odor representations in a widely projecting modulatory neuron
<p>This is the data set for Zhang et al. 2019 "Local synaptic inputs support opposing, network-specific odor representations in a widely projecting modulatory neuron" published at eLife.</p>
Dataset : Identifying locations susceptible to micro-anatomical reentry using a spatial network representation of atrial fibre maps
<ul> <li><strong>The three files in the dataset are:</strong></li> </ul> <p>1) Healthy Sheep Atria Fibre Orientation Dataset 300µm</p> <p>2) Heart Failure Sheep Atria Fibre Orientation Dataset 300µm</p> <p>3) Human Atria Fibre Orientation Dataset 330µm</p> <ul> <li><strong>Data is stored as numpy binary files. Given below is an example .py script to open the flat datasets:</strong></li> </ul> <p> import numpy as np<br> data = np.load("Human_330um.npy")</p> <ul> <li><strong>Volume and fibre orientation dataset stored in flat format as given below:</strong></li> </ul> <p> i, j, k, v1, v2, v3, ... repeated for each voxel </p> <p>where (i, j, k) are voxel coordinates and (v1, v2, v3) are vector components corresponding to fibre orientation within that voxel.</p>
Dataset1 for "General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian"
<p>Supporting data for the paper "General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian". </p> <p>Contains atomic structures and Hamiltonian matrices of monolayer graphene, monolayer MoS<sub>2</sub>, bilayer graphene and bilayer bismuthene.</p> <p>Detailed descriptions about the format of data and instructions on how to reproduce the results in the paper can be found in README.md.</p>
Dataset2 for "General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian"
<p>Supporting data for the paper "General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian". </p> <p>Contains atomic structures and Hamiltonian matrices of bilayer bismuth selenide. </p> <p>Detailed descriptions about the format of data and instructions on how to reproduce the results in the paper can be found in README.md.</p>
Dataset3 for "General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian"
<p>Supporting data for the paper "General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian". </p> <p>Contains atomic structures and Hamiltonian matrices of bilayer bismuth telluride. </p> <p>Detailed descriptions about the format of data and instructions on how to reproduce the results in the paper can be found in README.md.</p>
IoT network traffic dataset using the custom flow representation
Open the record for dataset details and reuse information.
PhyloCNN: Improving tree representation and neural network architecture for deep learning from trees in phylodynamics and diversification studies
Open the record for dataset details and reuse information.
Dataset for Utilizing Pathway Dependency and Network Representation of Anticancer Drug Sensitivity for Drug Synergy Prediction
Open the record for dataset details and reuse information.
Learning Useful Representations of Recurrent Neural Network Weight Matrices
<p>Dataset of RNN weights for the ICML 2024 paper "Learning Useful Representations of Recurrent Neural Network Weight Matrices". See <a title="GitHub repository" href="https://github.com/vincentherrmann/rnn-weights-representation-learning">https://github.com/vincentherrmann/rnn-weights-representation-learning</a>.</p> <h1> </h1>
Input data and scripts for "Spatial conservation prioritization for the East Asian islands: a balanced representation of multi-taxon biogeography in a protected area network"
<p>This release contains the input files 'input_data.zip' for the spatial conservation prioritization analysis by Zonation software, which are conducted in Lehtomäki et al. Input data includes biodiversity features (species distribution maps from vascular plants, mammals, birds, reptiles, amphibians, and freshwater fishes), habitat condition map (human influence index), priority mask information (the categorized protected area distribution) and the Japanese prefecture polygons in GeoTiff format, and the list of species attributes for conservation weighting in CSV format. Note that endangered rare species have been excluded from this dataset, though they were reflected in the output files. The Zonation setting files and R scripts for pre- and post analyses are included in 'japan-zsetup-1.0.zip' and also placed at GitHub : https://github.com/cbig/japan-zsetup</p> <p>The output files (priority score maps and removal curves) from the original Zonation analyses are summarized in 'output_from_original.data.zip'</p>
Data for "Convergent temperature representations in artificial and biological neural networks"
<p>Data for "Convergent Temperature Representations in Artificial and Biological<br> neural networks" by Haesemeyer M, Schier AF and Engert F, 2019</p> <p>The corresponding python code is available at:<br> <a href="https://github.com/haesemeyer/GradientPrediction">https://github.com/haesemeyer/GradientPrediction</a></p> <p>All zip files should be extracted in the same folder as the python files. This<br> will create a sub-folder structure for the model data.<br> ZIP File Contents (Note: These are used by the code and not necessarily useful<br> by themselves):<br> model_data.zip<br> Contains tensorflow checkpoints on all naive and fully trained models, test<br> errors during training as well as evolution weights where applicable.<br> model_cluster_assignments.zip<br> For the trained models in model_data.zip the response cluster assignment<br> for each individual unit.<br> zebrafish_data.zip<br> The zebrafish brain and behavior data used in the paper comparisons. This<br> archive also contains the temperature stimulus file stimFile.hdf5<br> training_data.zip<br> The generated training data used during predictive network training<br> test_data.zip<br> The generated test data used to evaluate predictive network training</p>
Subnetwork representation learning for discovering network biomarkers in predicting lymph node metastasis in early oral cancer
<p>Dataset for the specified study.</p>
Data and Codes of "A Pharmacological Representation-based LSTM Network for Drug–Drug Interaction Extraction"
<p>The datasets and source codes used and/or analyzed in study "A Pharmacological Representation-based LSTM Network for Drug–Drug Interaction Extraction".</p>
How biased is our perception of plant-pollinator networks? A comparison of visit- and pollen-based representations of the same networks
<p>These are the files (.txt) used to run the analyses in the paper "<em>How biased is our perception of plant-pollinator networks? A comparison of visit- and pollen-based representations of the same networks</em>" by de Manincor et al., 2020, Acta Oecologica, volume 105 (<a href="https://doi.org/10.1016/j.actao.2020.103551">https://doi.org/10.1016/j.actao.2020.103551</a>).</p> <p>The "netwoks_txt.zip" were used to run the motifs and positions analysis and the LBMs (+ alluvial).</p> <p>The "quilt.plot_txt.zip" were used to create the quiltplots (in the Supplementary materials of the paper).</p> <p>The "plant species list" provides information about the presence of plant species in the three sites/regions, and which species have been grouped at the genus level for the comparison between visits (obs) and pollen matrix.</p> <p> </p> <p><strong>ABSTRACT</strong></p> <p>Most plant-pollinator networks are based on observations of contact between an insect and a flower in the field. Despite significant sampling efforts, some links are easier to report, while others remain unobserved. Therefore, visit-based networks represent a subsample of possible interactions in which the ignored part is variable. Pollen is a natural marker of insect visits to flowers. The identification of pollen found on insect bodies can be used as an alternative method to study plant-pollinator interactions, with a potentially lower risk of bias than the observation of visits, since it increases the number of interactions in the network. Here we compare plant-pollinator networks constructed (i) from direct observation of pollinator visits and (ii) from identification of pollen found<br> on the same insects. We focused on three calcareous grasslands in France, with different plant and pollinator species diversities. Since pollen identification always yields richer, more connected networks, we focused our comparisons on sampling bias at equal network connectance. To do so, we first compared network structures with an analysis of latent blocks and motifs. We then compared species roles between both types of networks with an analysis of specialization and species positions within motifs. Our results suggest that the sampling from observations of insect visits does not lead to the construction of a network intrinsically different from the one obtained using pollen found on insect bodies, at least when field sampling strives to be exhaustive. Most of the<br> significant differences are found at the species level, not at the network structure level, with singleton species accounting for a respectable fraction of these differences. Overall, this suggests that recording plant-pollinator interactions from pollinator visit observation does not provide a biased picture of the network structure, regardless of species richness; however, it provided less information on species roles than the pollen-based network.</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.