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.
142
datasets available to search
ShareScore release 0.9.0
Dataset results
142 results for “zoos”
Galaxy Zoo DESI: Detailed Morphology Classifications for 8.7M Galaxies in the DESI Legacy Imaging Surveys
<p>This repository contains the data released in the paper "Galaxy Zoo DESI: Detailed Morphology Classifications for 8.7M Galaxies in the DESI Legacy Imaging Surveys" <em>(DOI to follow on publication).</em></p> <p>We release detailed morphology measurements for bright (<em>r </em>< 19) galaxies in the DESI Legacy Imaging Surveys footprint. These measurements estimate the presence of bars, spirals arms, ongoing mergers, and more.</p> <p>---</p> <p><strong>GZ DESI Detailed Morphology Catalogs</strong></p> <p>These catalogs are created by training deep learning models on Galaxy Zoo volunteer responses, to predict what volunteers might say for new galaxies. The models are available at [www.github.com/mwalmsley/zoobot](www.github.com/mwalmsley/zoobot). Our measurements are predicted vote fractions i.e. the fraction of volunteers expected to select a given answer for a given question.</p> <p>We share two catalog versions containing the same morphology measurements but presented in different ways.</p> <p>gz_desi_deep_learning_catalog_friendly.parquet contains the morphology measurements</p> <p>gz_desi_deep_learning_catalog_advanced.parquet contains the same measurements, and additional information:</p> <p>- _friendly includes only relevant vote fractions, defined as vote fractions to answers of questions that a majority of volunteers would have been asked. This removes predicted vote fractions for e.g. the fraction of volunteers answering "2 spiral arms" to a galaxy with no spiral arms. _advanced includes all vote fractions and instead reports the (column "proportion_asked"). The user must select which vote fractions they consider relevant (we suggest proportion_asked > 0.5, which recovers the _friendly fractions).</p> <p>- _advanced includes columns with estimated credible intervals (error bars) around each vote fraction. These are calculated from the vote fraction posterior predicted by our models.</p> <p>Finally, we separately present volunteer votes collected for 96k galaxies during the GZD-8 campaign, i.e. after the release of GZ DECaLS but before this (GZ DESI) release. These are split into the _core and _extended catalogs, where _extended includes galaxies which received five or more votes for "artifact". The models above were trained on these votes as well as votes from GZ DECaLS.</p> <p>---</p> <p><strong>External Catalog</strong></p> <p>For convenience, we also include an additional catalog of non-morphology measurements created by other authors (external_catalog.parquet) cross-matched to our morphology catalogs. Please credit those authors if you use this catalog (references are in the GZ DESI paper).</p> <p>A particularly important external measurement is redshift. Morphology is increasingly hard to resolve at higher redshift and so <strong>distant galaxies appear less featured</strong>. external_catalog.parquet includes the column "redshift", which is the SDSS spectroscopic redshift where available and a photometric redshift estimate otherwise (again, see the GZ DESI paper for references and credit). You may want to select only galaxies at lower redshifts.</p> <p>---</p> <p><strong>Data Notes</strong></p> <p>Parquet is a fast csv-like format which can be read with pd.read_parquet(loc, columns=[some columns]). Parquet files are read column-by-column (rather than row-by-row) and so you can chose which columns to load. You can easily check which columns are available using columns=['foo'] and reading the error message. We suggest loading only the columns you need when working with the larger catalogs. This will require much less memory than loading every column.</p> <p>We will release updates if needed via Zenodo versioning. We recommend using the latest version of this repository. You can check the version you are currently viewing on the right-hand sidebar.</p> <p>Please cite the paper (DOI to follow on publication) when using the data in this repository.</p> <p>---</p> <p><strong>History</strong></p> <p>v0.0.1 - closed pre-release for internal review</p> <p>v1.0.0 - draft public release. Removed low-z pre-filtered catalogs.</p> <p>v1.0.1 - first public release. Added .csv version of _friendly catalog. Tweaked catalog formatting for clarity and consistency.</p>
Fig. 5 in Renal trematode infection due to Paratanaisia bragai in zoo housed Columbiformes and a red bird-of-paradise (Paradisaea rubra)
Fig. 5. Allopeas clavulinum one of two subulinid snails possibly acting as intermediate host for the Paratanaisia bragai infection, found in one of the aviaries housing several of the cases.
Fig. 4 in Renal trematode infection due to Paratanaisia bragai in zoo housed Columbiformes and a red bird-of-paradise (Paradisaea rubra)
Fig. 4. Emerald dove case: Kidney with dilated collecting ducts and numerous cross section of trematodes (arrow) with minimal inflammation. HE.
Fig. 1 in Renal trematode infection due to Paratanaisia bragai in zoo housed Columbiformes and a red bird-of-paradise (Paradisaea rubra)
Fig. 1. Red bird-of-paradise case. (A) Renal flukes in collecting ducts (arrows) with minimal inflammatory changes. HE. (B) Necrosis and granulomatous nephritis surrounding trematode eggs (arrow). HE.
Fig. 3. Socorro dove case 2 in Renal trematode infection due to Paratanaisia bragai in zoo housed Columbiformes and a red bird-of-paradise (Paradisaea rubra)
Fig. 3. Socorro dove case 2. (A) Asymmetrically enlarged pale tan left and atrophic / cystic right kidney. Macroscopic view. (B) Kidney with focally extensive central necrosis surrounding fluke eggs (arrow) and surrounding granulomatous inflammatory reaction. HE. (C) Kidney with chronic granulomatous tubulointerstitial nephritis surrounding free trematode eggs. HE.
Fig. 2. Socorro dove case 1 in Renal trematode infection due to Paratanaisia bragai in zoo housed Columbiformes and a red bird-of-paradise (Paradisaea rubra)
Fig. 2. Socorro dove case 1. (A) Kidneys with cranial atrophy, haemorrhage and caudal polar gout deposition and also pericardial gout deposition. Macroscopic view. (B) Cystic dilation of collecting ducts, haemorrhage and granulomatous nephritis. HE.
Model Zoo: A Dataset of Diverse Populations of Resnet-18 Models - CIFAR-10
<p><strong>Abstract</strong></p> <p>In the last years, neural networks have evolved from laboratory environments to the state-of-the-art for many real-world problems. Our hypothesis is that neural network models (i.e., their weights and biases) evolve on unique, smooth trajectories in weight space during training. Following, a population of such neural network models (refereed to as “model zoo”) would form topological structures in weight space. We think that the geometry, curvature and smoothness of these structures contain information about the state of training and can be reveal latent properties of individual models. With such zoos, one could investigate novel approaches for (i) model analysis, (ii) discover unknown learning dynamics, (iii) learn rich representations of such populations, or (iv) exploit the model zoos for generative modelling of neural network weights and biases. Unfortunately, the lack of standardized model zoos and available benchmarks significantly increases the friction for further research about populations of neural networks. With this work, we publish a novel dataset of model zoos containing systematically generated and diverse populations of neural network models for further research. In total the proposed model zoo dataset is based on six image datasets, consist of 24 model zoos with varying hyperparameter combinations are generated and includes 47’360 unique neural network models resulting in over 2’415’360 collected model states. Additionally, to the model zoo data we provide an in-depth analysis of the zoos and provide benchmarks for multiple downstream tasks as mentioned before.</p> <p><strong>Dataset</strong></p> <p>This dataset is part of a larger collection of model zoos and contains the zoo of 1000 ResNet18 models trained on CIFAR10. All zoos with extensive information and code can be found at www.modelzoos.cc.</p> <p>The complete zoo is 2.3TB large. Due to the size, this repository contains the checkpoints of epochs 1, 10 and 50. For a link to the full dataset as well as more information on the zoos and code to access and use the zoos, please see www.modelzoos.cc.</p> <p> </p>
Model Zoo: A Dataset of Diverse Populations of Resnet-18 Models - CIFAR-100
<p><strong>Abstract</strong></p> <p>In the last years, neural networks have evolved from laboratory environments to the state-of-the-art for many real-world problems. Our hypothesis is that neural network models (i.e., their weights and biases) evolve on unique, smooth trajectories in weight space during training. Following, a population of such neural network models (refereed to as “model zoo”) would form topological structures in weight space. We think that the geometry, curvature and smoothness of these structures contain information about the state of training and can be reveal latent properties of individual models. With such zoos, one could investigate novel approaches for (i) model analysis, (ii) discover unknown learning dynamics, (iii) learn rich representations of such populations, or (iv) exploit the model zoos for generative modelling of neural network weights and biases. Unfortunately, the lack of standardized model zoos and available benchmarks significantly increases the friction for further research about populations of neural networks. With this work, we publish a novel dataset of model zoos containing systematically generated and diverse populations of neural network models for further research. In total the proposed model zoo dataset is based on six image datasets, consist of 27 model zoos with varying hyperparameter combinations are generated and includes 50’360 unique neural network models resulting in over 2’585’360 collected model states. Additionally, to the model zoo data we provide an in-depth analysis of the zoos and provide benchmarks for multiple downstream tasks as mentioned before.</p> <p><strong>Dataset</strong></p> <p>This dataset is part of a larger collection of model zoos and contains the zoo of 1000 ResNet18 models trained on CIFAR100. All zoos with extensive information and code can be found at www.modelzoos.cc.</p> <p>The complete zoo is 2.6TB large. Due to the size, this repository contains the checkpoints of the last epoch 60. For a link to the full dataset as well as more information on the zoos and code to access and use the zoos, please see www.modelzoos.cc.</p>
Sparsified Model Zoo Twins: A Dataset of Sparsified Populations of Neural Network Models - MNIST
<p><strong>Abstract</strong></p> <p>In the last years, neural networks have evolved from laboratory environments to the state-of-the-art for many real-world problems. Our hypothesis is that neural network models (i.e., their weights and biases) evolve on unique, smooth trajectories in weight space during training. Following, a population of such neural network models (refereed to as “model zoo”) would form topological structures in weight space. We think that the geometry, curvature and smoothness of these structures contain information about the state of training and can be reveal latent properties of individual models. With such zoos, one could investigate novel approaches for (i) model analysis, (ii) discover unknown learning dynamics, (iii) learn rich representations of such populations, or (iv) exploit the model zoos for generative modelling of neural network weights and biases. Unfortunately, the lack of standardized model zoos and available benchmarks significantly increases the friction for further research about populations of neural networks. With this work, we publish a novel dataset of model zoos containing systematically generated and diverse populations of neural network models for further research. In total the proposed model zoo dataset is based on six image datasets, consist of 27 model zoos with varying hyperparameter combinations are generated and includes 50’360 unique neural network models resulting in over 2’585’360 collected model states. Additionally, to the model zoo data we provide an in-depth analysis of the zoos and provide benchmarks for multiple downstream tasks as mentioned before.</p> <p><strong>Dataset</strong></p> <p>This dataset is part of a larger collection of model zoos and contains the sparsified twins of models trained on MNIST. The original population is made available at https://doi.org/10.5281/zenodo.6632086. Sparsification is done using Variational Dropout, starting from the last epoch of the original population. The zip file contains the sparsification trajectory for 25 epochs for all 1000 models. All zoos with extensive information and code can be found at www.modelzoos.cc.</p> <p> </p>
Model Zoo: A Dataset of Diverse Populations of Resnet-18 Models - Tiny ImageNet
<p><strong>Abstract</strong></p> <p>In the last years, neural networks have evolved from laboratory environments to the state-of-the-art for many real-world problems. Our hypothesis is that neural network models (i.e., their weights and biases) evolve on unique, smooth trajectories in weight space during training. Following, a population of such neural network models (refereed to as “model zoo”) would form topological structures in weight space. We think that the geometry, curvature and smoothness of these structures contain information about the state of training and can be reveal latent properties of individual models. With such zoos, one could investigate novel approaches for (i) model analysis, (ii) discover unknown learning dynamics, (iii) learn rich representations of such populations, or (iv) exploit the model zoos for generative modelling of neural network weights and biases. Unfortunately, the lack of standardized model zoos and available benchmarks significantly increases the friction for further research about populations of neural networks. With this work, we publish a novel dataset of model zoos containing systematically generated and diverse populations of neural network models for further research. In total the proposed model zoo dataset is based on six image datasets, consist of 27 model zoos with varying hyperparameter combinations are generated and includes 50’360 unique neural network models resulting in over 2’585’360 collected model states. Additionally, to the model zoo data we provide an in-depth analysis of the zoos and provide benchmarks for multiple downstream tasks as mentioned before.</p> <p><strong>Dataset</strong></p> <p>This dataset is part of a larger collection of model zoos and contains the zoo of 1000 ResNet18 models trained on Tiny Imagenet. All zoos with extensive information and code can be found at www.modelzoos.cc.</p> <p>The complete zoo is 2.6TB large. Due to the size, this repository contains the checkpoints of the first 115 models at their last epoch 60. For a link to the full dataset as well as more information on the zoos and code to access and use the zoos, please see www.modelzoos.cc.</p>
Sparsified Model Zoo Twins: A Dataset of Sparsified Populations of Neural Network Models - SVHN
<p><strong>Abstract</strong></p> <p>In the last years, neural networks have evolved from laboratory environments to the state-of-the-art for many real-world problems. Our hypothesis is that neural network models (i.e., their weights and biases) evolve on unique, smooth trajectories in weight space during training. Following, a population of such neural network models (refereed to as “model zoo”) would form topological structures in weight space. We think that the geometry, curvature and smoothness of these structures contain information about the state of training and can be reveal latent properties of individual models. With such zoos, one could investigate novel approaches for (i) model analysis, (ii) discover unknown learning dynamics, (iii) learn rich representations of such populations, or (iv) exploit the model zoos for generative modelling of neural network weights and biases. Unfortunately, the lack of standardized model zoos and available benchmarks significantly increases the friction for further research about populations of neural networks. With this work, we publish a novel dataset of model zoos containing systematically generated and diverse populations of neural network models for further research. In total the proposed model zoo dataset is based on six image datasets, consist of 27 model zoos with varying hyperparameter combinations are generated and includes 50’360 unique neural network models resulting in over 2’585’360 collected model states. Additionally, to the model zoo data we provide an in-depth analysis of the zoos and provide benchmarks for multiple downstream tasks as mentioned before.</p> <p><strong>Dataset</strong></p> <p>This dataset is part of a larger collection of model zoos and contains the sparsified twins of models trained on SVHN. The original population is made available at https://doi.org/10.5281/zenodo.6632120. Sparsification is done using Variational Dropout, starting from the last epoch of the original population. The zip file contains the sparsification trajectory for 25 epochs for all 1000 models. All zoos with extensive information and code can be found at www.modelzoos.cc.</p>
MBW Zoo Data
<p>We release a challenging dataset consisting image frames of tail-end distribution categories (such as Fish, Colobus Monkeys, Chimpanzees, etc.) with their corresponding 2D, 3D, and Bounding-Box labels generated from minimal human intervention. Some of the prominent use cases of this dataset include not only sparse <strong>2D and 3D landmark prediction</strong>, but also dense reconstruction tasks such as <strong>dense deformable shape reconstruction</strong>, novel view rendering (<strong>NeRF</strong>), <strong>Detection and Tracking,</strong> and finally this dataset could also be used for camera estimation in Simultaneous Localization and Mapping (<strong>SLAM</strong>) frameworks.</p>
Model Zoo for Robust Models are less Over-Confident
<p><strong>Model Zoo (PyTorch) of non-adversarially trained models for Robust Models are less Over-Confident (NeurIPS'22)</strong></p> <p>Abstract: <em>"Regardless of the success of convolutional neural networks (CNNs) in many academic benchmarks of computer vision tasks, their application in real-world is still facing fundamental challenges, like the inherent lack of robustness as unveiled by adversarial attacks. These attacks target to manipulate the network's prediction by adding a small amount of noise onto the input. In turn, adversarial training (AT) aims to achieve robustness against such attacks by including adversarial samples in the trainingset. However, a general analysis of the reliability and model calibration of these robust models beyond adversarial robustness is still pending. In this paper, we analyze a variety of adversarially trained models that achieve high robust accuracies when facing state-of-the-art attacks and we show that AT has an interesting side-effect: it leads to models that are significantly less overconfident with their decisions even on clean data than non-robust models. Further, our analysis of robust models shows that not only AT but also the model's building blocks (activation functions and pooling) have a strong influence on the models' confidence."</em></p>
Transformers Model Zoos and Soups: A Population of Language and Vision Models
<p>Model Zoos submitted to the NeurIPS 2024 Dataset & Benchmark track: "<em>Transformer Model Zoos and Soups: A Population of Language and Vision Models</em>"</p> <p>We generate two model zoos, one for computer vision built on the ViT-S architecture, and one for language modeling based on the BERT architecture. For each, we train several backbone models with varying hyperparameters, and further fine-tune them using multiple hyperparameter combinations. We further annotate every model with performance metrics. These include test accuracy and F1-score, as well as the generalization gap. For the vision models, we also include the robust accuracy after a FGSM attack.</p>
Figure 3 in Toxoplasmosis in a bar-shouldered dove (Geopelia humeralis) from the Zoo of Clères, France
Figure 3. Toxoplasma gondii in a bar-shouldered dove, lung. Two tachyzoites enclosed in a parasitophorous vacuolar membrane (pvm). Note conoid (co), micronemes (mn), rhoptries (ro) with honey-combed contents, and a nucleus (nu) in each tachyzoite. The parasitophorous vacuole has membranous tubules. Transmission electron microscopy.
Figure 2 in Toxoplasmosis in a bar-shouldered dove (Geopelia humeralis) from the Zoo of Clères, France
Figure 2. Toxoplasma gondii in a bar-shouldered dove, lung. Note a tachyzoite with dividing nucleus (arrow) and individual tachyzoites (arrowheads). H&E stain.
Figure 1 in Toxoplasmosis in a bar-shouldered dove (Geopelia humeralis) from the Zoo of Clères, France
Figure 1. Toxoplasma gondii in a bar-shouldered dove, lung. Note the necrotic process. There are numerous intralesional tachyzoites but non-visible at this magnification. H&E stain.
Fig. 2 in Plasmodium (Novyella) nucleophilum from an Egyptian Goose in São Paulo Zoo, Brazil: microscopic confirmation and molecular characterization
Fig. 2. Bayesian phylogeny of cytochrome b gene lineages of species of avian haemosporidian parasites. A lineage recorded in the Egyptian Goose Alopochen aegyptiacus is provided underlined. Names of the lineages are given after the species names of parasites. GenBank accession numbers of the lineages are provided before the parasite species names. Nodal support values (in percentage) indicate posterior clade probabilities. Plasmodium species from Novyella subgenus are boxed.
Fig. 1 in Plasmodium (Novyella) nucleophilum from an Egyptian Goose in São Paulo Zoo, Brazil: microscopic confirmation and molecular characterization
Fig. 1. Photomicrographs of Plasmodium parasites visualized from thin blood smears obtained from an Egyptian Goose (Alopochen aegyptiacus) in São Paulo Zoo, Brazil. Characteristic of Plasmodium (Novyella) nucleophilum (lineage EG01, GenBank JX467689) the trophozoite (a), meronts (b–c), macrogametocytes (d, e), and microgametocyte (f) are appressed to erythrocyte nuclei (nucleophilic features). Plasmodium (Haemamoeba) sp. (g–i) lacks nucleophilic blood stages and possesses large roundish trophozoites, each with a prominent centrally located vacuole; pigment granules are gathered around the vacuoles. Note that early Plasmodium (H.) sp. trophozoites markedly displace erythrocyte nuclei (g). Arrows, pigment granules. Scale bar = 10 µm.
Fig. 1 in Molecular characterization of Blastocystis sp. in captive wildlife in Bangladesh National Zoo: Non-human primates with high prevalence and zoonotic significance
Fig. 1. Phylogenetic tree of the Blastocystis sp. isolates and reference SSU rRNA gene sequences from GenBank based on maximum likelihood analysis. The tree was rooted on Karotomorpha sp. and Protoopalina intestinalis. Bootstrap values> 50% from 1,000 replicates are shown on the nodes. Reference sequences from GenBank have accession number and host designation. The isolates of seven subtypes, with their host designations, are indicated by triangle shape.
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.