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.
99
datasets available to search
ShareScore release 0.9.0
Dataset results
99 results for “Transfer learning”
Accurate TCR-pMHC Interaction Prediction Using a BERT-based Transfer Learning Method
<p>The datasets used for the TABR-BERT. For the complete training and testing code of TABR-BERT, see <a href="https://github.com/Freshwind-Bioinformatics/TABR-BERT">Freshwind-Bioinformatics/TABR-BERT: TABR-BERT: an Accurate and Robust BERT-based Transfer Learning Model for TCR-pMHC Interaction Prediction (github.com)</a>.</p>
Data from: Generalizable physical descriptors of pool boiling heat transfer from unsupervised learning of images
Open the record for dataset details and reuse information.
Transfer learning enables prediction of CYP2D6 haplotype function
<p>This data here were used to train the models described in the manuscript "Transfer learning enables prediction of CYP2D6 haplotype function". The deep learning model described predicts metabolic function of <em>CYP2D6</em> star alleles. It uses two pretraining steps, first with simulated data, then with sequence data collected from liver microsomes, and finally using sequence data for <em>CYP2D6</em> star alleles.</p> <p> </p> <p>simulated_cyp2d6_diplotypes.tar.gz - This file contains sequence data and labels for simulated <em>CYP2D6 </em>data used in the first training step</p> <p>dalton_2019_cyp2d6_microsomes.txt - This file contains summary statistic data for liver microsome data used in the second pretraining step (originally from <a href="https://doi.org/10.1111/cts.12695">https://doi.org/10.1111/cts.12695)</a></p> <p>star_samples.vcf - This file contains sequence data for <em>CYP2D6 </em>star alleles derived from PharmVar (https://www.pharmvar.org/gene/CYP2D6) used in the final training step.</p>
Transfer Learning Models and Datasets for a Reliable Emergency Landing Field Identification
<p>The file <em>data.tar.gz</em> compromises three HDF5 datasets. This file has been split into 100 files. The files can be merged, decompressed and unpacked with the following commands:</p> <pre><code class="language-bash">cat data* > data.tar.gz tar -xzf data.tar.gz</code></pre> <p>Afterwards, the three files: <em>train_test_data_ss8_supervised_new.hdf5</em>, <em>train_test_data_ss16_supervised_new.hdf5</em>, <em>train_test_data_ss32_supervised_new.hdf5</em> are ready to get processed. </p> <p>Internal structure of the datasets:<br> <strong>Search Window (SW) 8 m^2:</strong><br> HDF5 "train_test_data_ss8_supervised_new.hdf5" {<br> GROUP "/" {<br> GROUP "test" {<br> DATASET "fus_data" {<br> DATATYPE H5T_IEEE_F32LE<br> DATASPACE SIMPLE { ( 76288, 40, 40, 8 ) / ( 76382, 40, 40, 8 ) }<br> }<br> DATASET "labels" {<br> DATATYPE H5T_STD_I64LE<br> DATASPACE SIMPLE { ( 76382, 1 ) / ( 76382, 1 ) }<br> }<br> }<br> GROUP "train" {<br> DATASET "fus_data" {<br> DATATYPE H5T_IEEE_F32LE<br> DATASPACE SIMPLE { ( 380928, 40, 40, 8 ) / ( 380998, 40, 40, 8 ) }<br> }<br> DATASET "labels" {<br> DATATYPE H5T_STD_I64LE<br> DATASPACE SIMPLE { ( 380998, 1 ) / ( 380998, 1 ) }<br> }<br> }<br> }}</p> <p><br> <strong>SW 16 m^2:</strong><br> HDF5 "train_test_data_ss16_supervised_new.hdf5" {<br> GROUP "/" {<br> GROUP "test" {<br> DATASET "fus_data" {<br> DATATYPE H5T_IEEE_F32LE<br> DATASPACE SIMPLE { ( 17024, 80, 80, 8 ) / ( 17054, 80, 80, 8 ) }<br> }<br> DATASET "labels" {<br> DATATYPE H5T_STD_I64LE<br> DATASPACE SIMPLE { ( 17054, 1 ) / ( 17054, 1 ) }<br> }<br> }<br> GROUP "train" {<br> DATASET "fus_data" {<br> DATATYPE H5T_IEEE_F32LE<br> DATASPACE SIMPLE { ( 84992, 80, 80, 8 ) / ( 85068, 80, 80, 8 ) }<br> }<br> DATASET "labels" {<br> DATATYPE H5T_STD_I64LE<br> DATASPACE SIMPLE { ( 85068, 1 ) / ( 85068, 1 ) }<br> }<br> }<br> }}</p> <p><br> <strong>SW 32 m^2:</strong><br> HDF5 "train_test_data_ss32_supervised_new.hdf5" {<br> GROUP "/" {<br> GROUP "test" {<br> DATASET "fus_data" {<br> DATATYPE H5T_IEEE_F32LE<br> DATASPACE SIMPLE { ( 3328, 160, 160, 8 ) / ( 3359, 160, 160, 8 ) }<br> }<br> DATASET "labels" {<br> DATATYPE H5T_STD_I64LE<br> DATASPACE SIMPLE { ( 3359, 1 ) / ( 3359, 1 ) }<br> }<br> }<br> GROUP "train" {<br> DATASET "fus_data" {<br> DATATYPE H5T_IEEE_F32LE<br> DATASPACE SIMPLE { ( 16768, 160, 160, 8 ) / ( 16793, 160, 160, 8 ) }<br> }<br> DATASET "labels" {<br> DATATYPE H5T_STD_I64LE<br> DATASPACE SIMPLE { ( 16793, 1 ) / ( 16793, 1 ) }<br> }<br> }<br> }}</p> <p>The sample count of the various generated dataset is as follows: <br> <strong>SW 8 m^2:</strong> {train: 380,928 with {0: 190,464, 1: 190,464}, test: 76,288 with {0: 38,152, 1: 38,136}}<br> <strong>SW 16 m^2: </strong>{train: 84,992 with {0: 42,498, 1: 42,494}, test: 17,024 with {0: 8,516, 1: 8,508}}<br> <strong>SW 32 m^2: </strong>{train: 16,768 with {0: 8,424, 1: 8,344}, test: 3,328 with {0: 1,672, 1: 1,656}}</p> <p>Each sample is composed as follows:<br> RGB = sample[:,:,<strong>:3</strong>]; Theoretically: [0, 1] per color channel<br> NIR = sample[:,:,<strong>3</strong>]; Theoretically: [0, 1]<br> Slope = sample[:,:,<strong>4</strong>]; Theoretically: [0, 90]<br> Roughness = sample[:,:,<strong>5</strong>]; Theoretically: [0, 78.78]<br> NDVI = sample[:,:,<strong>6</strong>]; Theoretically: [-1, 1]<br> DOM = sample[:,:,<strong>7</strong>]; Theoretically: [0, 429.90]</p> <p>====================================================================================================</p> <p>The following three files compromise the model and optimizer state variable of our PyTorch models trained on the aforementioned datasets: <em>best_alexnet_final.pth</em>, <em>best_resnet18_final.pth</em>, <em>best_wide_resnet50_2_final.pth</em></p> <p>Below find a more precise description of each model:<br> <strong>best_resnet18_final.pth</strong></p> <ul> <li>Model: ResNet-18</li> <li>Dataset: SW 8</li> <li>Input: RGB-NIR-Slope -> R: [0,224,224], G: [1,224,224], B: [2,224,224], NIR: [3,224,224], Slope: [4,224,224]</li> </ul> <p><strong>best_wide_resnet50_2_final.pth</strong></p> <ul> <li>Model: Wide-ResNet-50-2</li> <li>Dataset: SW 16</li> <li>Input: NDVI-Slope -> NDVI: [0,224,224], Slope: [1,224,224]</li> </ul> <p><strong>best_alexnet_final.pth</strong></p> <ul> <li>Model: AlexNet</li> <li>Dataset: SW 32</li> <li>Input: RGB-Slope -> R: [0,224,224], G: [1,224,224], B: [2,224,224], Slope: [3,224,224]</li> </ul> <p>Each model is capable of performing a binary classification, distinguishing between landable and unlandable samples</p>
Data-driven subgrid-scale modeling of forced Burgers turbulence using deep learning with generalization to higher Reynolds numbers via transfer learning
<p>These are the data files for use with the codes in https://github.com/envfluids/Burgers_DDP_and_TL.</p>
Large-scale semantic indexing of Spanish biomedical literature using contrastive transfer learning
Open the record for dataset details and reuse information.
MOTL: enhancing multi-omics matrix factorization with transfer learning
<p>The <strong>Lrn_5000D_Fctrzn_100k_001TH.zip</strong> file contains the results of a MOFA factorization of the TGCA learning dataset, to be downloaded and used for transfer learning factorization of a target dataset with MOTL. The MOFA output is in the <strong>Model.hdf5</strong> file, and intercepts for the factorization are in the <strong>EstimatedIntercepts.rds</strong> file. The <strong>FctrMeta.json </strong>file contains metadata related to the MOFA factorization. The nohup.out file is the log of the factorization.</p> <p>The <strong>expdat_meta.rds</strong> file contains metadata from the preprocessing of the TCGA multi-omics learning dataset that was factorized. This should also to be downloaded as it is an input to MOTL</p>
Machine learning models predict calculation outcomes with the transferability necessary for computational catalysis
<p>data files, including ML models of dynamic classifiers, trajectories of electronic structure and geometric features, optimized geometries, and final csv files.</p>
Machine Learning Potentials for Metal-Organic Frameworks with Thermodynamic Transferability: training data
<p>This dataset contains potential energies, forces, and virial stress for a large set of reference configurations for UiO-66(Zr) and MIL-53(Al), computed at the PBE-D3 level using CP2K 7.1. The basis set contained both TZVP Gaussian basis functions as well as plane waves (cutoff energy 800 Ry for UiO-66(Zr) and 900 Ry for MIL-53(Al)). The sampling of the Brillouin zone was restricted to the gamma point.</p>
Explaining the physics of transfer learning a data-driven subgrid-scale closure to a different turbulent flow
<p>Data for the six test cases of 2D turbulence explored in the paper, Explaining the physics of transfer learning a data-driven subgrid-scale closure to a different turbulent flow.</p>
Quantum Hardware-Enabled Molecular Dynamics via Transfer Learning
<p>This zip file contains datasets for training Behler-Parrinello neural networks using the n2p2 software. Both DFT and VQE training sets are given for both the water monomer and dimer dataset.</p>
GIST: Generated Inputs Sets Transferability in Deep Learning (Part 2)
<p>Part2 of the Replication Package for the paper "GIST: Generated Inputs Sets Transferability in Deep Learning"</p> <p>Contains RoBERTa models and data for the KMNC property.</p> <p>Github link: https://github.com/FlowSs/GIST</p> <p>Part1 can be found here: https://zenodo.org/records/10028594</p> <p>Abstract:</p> <p> To foster the verifiability and testability of Deep Neural Networks (DNN), an increasing number of methods<br>for test case generation techniques are being developed.<br> When confronted with testing DNN models, the user can apply any existing test generation technique.<br>However, it needs to do so for each technique and each DNN model under test, which can be expensive.<br>Therefore, a paradigm shift could benefit this testing process: rather than regenerating the test set independently<br>for each DNN model under test, we could transfer from existing DNN models.<br> This paper introduces GIST (Generated Inputs Sets Transferability), a novel approach for the efficient<br>transfer of test sets. Given a property selected by a user (e.g., neurons covered, faults), GIST enables the<br>selection of good test sets from the point of view of this property among available test sets. This allows the<br>user to recover similar properties on the transferred test sets as he would have obtained by generating the<br>test set from scratch with a test cases generation technique. Experimental results show that GIST can select<br>effective test sets for the given property to transfer. Moreover, GIST scales better than reapplying test case<br>generation techniques from scratch on DNN models under test.</p>
Nondestructive Fatigue Life Prediction for Additively Manufactured Metal Parts through a Multimodal Transfer Learning Framework
Open the record for dataset details and reuse information.
Deep Ensemble Learning and Transfer Learning Methods for Classification of Senescent Cells from Nonlinear Optical Microscopy Images
<p>This Dataset contains the train and test NLO images in pickle format used for the following publication: Deep Ensemble Learning and Transfer Learning Methods for Classification of Senescent Cells from Nonlinear Optical Microscopy Images</p>
A federated learning framework based on transfer learning and knowledge distillation for targeted advertising-Click-Through Rate Prediction Dataset
<p>https://www.kaggle.com/c/avazu-ctr-prediction</p>
Using 'Guided-Discovery-Learning' to Optimize and Maximize Transfer of Surgical Simulation
ClinicalTrials.gov study NCT03684720. IPD Sharing: NO. Countries: 1. Publications: 13.
Transfer Learning of a Neural Network for Robotic Surgical Assessment
ClinicalTrials.gov study NCT06612606. IPD Sharing: YES. Countries: 1. Publications: 0.
Bladder Cancer Staging and Prediction of New Adjuvant Chemotherapy Efficacy Based on Deep Learning and Transfer Learning in Ultrasound-Magnetic Resonance-Pathology Multimodal Multiscale
ClinicalTrials.gov study NCT07051083. IPD Sharing: YES. Countries: 1. Publications: 13.
Data from: From strategy to action: A qualitative study on salient factors influencing knowledge transfer in project-based experiential learning in healthcare organizations in Kenya
Open the record for dataset details and reuse information.
Transfer Learning for leveraging computer vision in infrastructure maintenance [extracted features]
<p>Dataset containing features extracted from images taken on single case of infrastructure facility for the purpose of training Transfer Learned CNN classifier. It is meant to be used with KrakN framework (https://github.com/MatZar01/KrakN), published with the research paper.</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.