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.
185
datasets available to search
ShareScore release 0.9.0
Dataset results
185 results for “fine-tuning”
CodonTransformer - Genomic and CodonTransformer-Generated Sequences for Fine-tuned Organisms
<p>This dataset is used in creating Fig. 2a and Supplementary Figs. 2-16 of the paper, mainly including the predictions of base (pretrained) and finetuend CodonTransformer model along with various metrics. </p>
Data from: Fine-tuning biodiversity assessments: A framework to pair eDNA metabarcoding and morphological approaches
<p><span>Accurate quantification of biodiversity can be demanding and expensive. Although environmental DNA (eDNA) metabarcoding can facilitate biodiversity assessments through non-invasive, cost-efficient, and rapid surveys, the approach struggles to outperform traditional morphological approaches in providing reliable quantitative estimates for surveyed species (e.g., abundance and biomass).</span></p> <p><span>We present an integrated methodology for improving biodiversity surveys that pairs eDNA metabarcoding with morphological data, following a series of taxonomic and geographic filters. We demonstrate its power by applying it to a new spatiotemporal dataset generated on an Iberian-wide distributed aquatic mesocosm infrastructure that spans a wide biogeographic gradient.</span></p> <p><span>By building upon the strengths that these two approaches offer, our framework improved taxonomic resolution for 30% of the taxa and enabled species' traits (e.g., body-size) and abundance to be assigned to 85% of the taxa in hybrid datasets.</span></p> <p><span>These results indicate that eDNA-based assessments can complement, but not always replace, conventional approaches. Integrating conventional and modern eDNA metabarcoding approaches, already available in the ecologist's toolbox, will greatly enhance biodiversity assessments.</span></p>
The impacts of fine-tuning, phylogenetic distance, and sample size on big-data bioacoustics
<p>Vocalizations in animals, particularly birds, are critically important behaviors that influence their reproductive fitness. While recordings of bioacoustic data have been captured and stored in collections for decades, the automated extraction of data from these recordings has only recently been facilitated by artificial intelligence methods. These have yet to be evaluated with respect to accuracy of different automation strategies and features. Here, we use a recently published machine learning framework to extract syllables from ten bird species ranging in their phylogenetic relatedness from 1 to 85 million years, to compare how phylogenetic relatedness influences accuracy. We also evaluate the utility of applying trained models to novel species. Our results indicate that model performance is best on conspecifics, with accuracy progressively decreasing as phylogenetic distance increases between taxa. However, we also find that the application of models trained on multiple distantly related species can improve the overall accuracy to levels near that of training and analyzing a model on the same species. When planning big-data bioacoustics studies, care must be taken in sample design to maximize sample size and minimize human labor without sacrificing accuracy.</p>
Data from: Fine-tuning the nested structure of pollination networks by adaptive interaction switching, biogeography and sampling effect in the Galápagos Islands
Open the record for dataset details and reuse information.
Supplementary material data for: Unstable environmental conditions constrain the fine-tune between opsin sensitivity and underwater light in an Amazon forest stream fish
Open the record for dataset details and reuse information.
Data from: High-fidelity parameter-efficient fine-tuning for joint recognition and linking of diagnoses to ICD-10 in non-standard primary care notes
Open the record for dataset details and reuse information.
The impacts of fine-tuning, phylogenetic distance, and sample size on big-data bioacoustics
Open the record for dataset details and reuse information.
Data from: Fine-tuning biodiversity assessments: A framework to pair eDNA metabarcoding and morphological approaches
Open the record for dataset details and reuse information.
Data from: Causes and consequences of repeatability, flexibility and individual fine-tuning of migratory timing in pike
1. Many organisms undertake migrations between foraging and breeding habitats and while it is assumed that reproductive timing affects fitness, little is known about the degree of individual consistency, and about the causes and consequences of individual variation in migratory timing in organisms other than birds. 2. Here, we report on a 6-year mark–recapture study, including 2048 individuals, of breeding migration in anadromous pike (Esox lucius), an iteroparous top-predatory fish that displays homing behaviour. By repeated sampling across years at a breeding site, we first quantify individual variation both within and between breeding events and then investigate phenotypic correlates and fitness consequences of arrival timing to the breeding site. 3. Our data demonstrate that males arrive before females, that large males arrive later than small males, that the timing of breeding migration varies among years and that individuals are consistent in their timing across years relative to other individuals in the population. 4. Furthermore, data on return rates indicate that arrival time is under stabilizing viability selection, and that individuals who are more flexible in their timing of arrival during the first reproductive years survive longer compared with less flexible individuals. Finally, longitudinal data demonstrate that individuals consistently fine-tune their arrival timing across years, showing that the timing of arrival to breeding sites is influenced by experience. 5. These findings represent rare evidence of how between- and within-individual variations in migratory timing across breeding events are correlated with phenotypic and fitness traits in an ecologically important keystone species. Our results emphasize the importance of considering variation in migratory timing both between and within individuals in studies investigating the fitness consequences of migratory behaviour and have implications for future management.
Pre-training and fine-tuning dataset for transformers consisting of basic blocks and their execution times (average, minimum, and maximum) along with the execution context of these blocks, for various Cortex processors M7, M4, A53, and A72.
<p>We are making public the dataset used for training CAWET, a tool for estimating the Worst-Case Execution Time (WCET) of basic blocks using the Transformer XL model. CAWET leverages the Transformer architecture for accurate WCET predictions, and its training involves two main phases: self-supervised pre-training and fine-tuning.</p><p>CAWET undergoes a pre-training process on a substantial corpus of basic blocks to enable the Transformer to grasp the intricacies of the assembly language in focus. For this, we utilized CodeNet \cite{codenet}, a comprehensive collection of publicly submitted solutions to competitive programming challenges, comprising roughly 900,000 C programs. These programs were cross-compiled to the target architecture and subsequently disassembled using GNU binary utilities with objdump. The textual output from objdump, post a series of basic parsing operations (e.g., address extraction, separation of basic blocks), serves as the foundation for an extensive pre-training dataset. We employed this dataset to develop a vocabulary model utilizing sentence piece \cite{sentencepiece}. Following the completion of the sentence piece model's training, it becomes ready for use in tokenizing any binary programs written in the target instruction set.</p><p>The fine-tuning phase of CAWET involves its adaptation to basic blocks along with their contextual information. Here, we used a varied and openly accessible collection of programs, namely, The Algorithms (accessible at: <a href="https://github.com/TheAlgorithms/C">https://github.com/TheAlgorithms/C</a>), MiBench \cite{mibench}, and Polybench \cite{polybench}.</p><p>The provided zip file encompasses the following directories:</p><p>Fine_Tuning: This includes four distinct files, each tailored for a specific processor: Cortex_M4, Cortex_M7, Cortex_A53, and Cortex_72. Each file encompasses the basic block under analysis (bbUA), the preceding 10 basic blocks executed prior to it, and timing information related to the bbUA (mean, min, max, normalization, etc.).</p><p>Pre_Training: This comprises two extensive files, dataset_CortexA and dataset_CortexM, utilized for pre-training the transformers for the Masked Language Modeling Task (MLM). Additionally, it includes the sentence piece model and the necessary code to facilitate accurate tokenization.</p><p>For additional information, please refer to the CAWET paper or contact us at <a href="mailto:ea_amalou@esi.dz">ea_amalou@esi.dz</a></p><p> </p><p>Citation:</p><p>@inproceedings{amalou2023cawet,</p><p> title={CAWET: Context-Aware Worst-Case Execution Time Estimation Using Transformers},</p><p> author={Amalou, Abderaouf N and Fromont, Elisa and Puaut, Isabelle},</p><p> booktitle={35th Euromicro Conference on Real-Time Systems (ECRTS 2023)},</p><p> year={2023},</p><p> organization={Schloss Dagstuhl-Leibniz-Zentrum f{\"u}r Informatik}</p><p>}</p><p> </p><p><strong>Bibliography</strong>:</p><p>codenet</p><p>@article{codenet2021,</p><p> title={CodeNet: A large-scale AI for code dataset for learning a diversity of coding tasks},</p><p> author={Puri, Ruchir and Kung, David S and Janssen, Geert and Zhang, Wei and Domeniconi, Giacomo and Zolotov, Vladimir and Dolby, Julian and Chen, Jie and Choudhury, Mihir and Decker, Lindsey and others},</p><p> journal={arXiv preprint arXiv:2105.12655},</p><p> year={2021}</p><p>}</p><p>sentencepiece</p><p>@article{sentencepiece2018,</p><p> title={Sentencepiece: A simple and language independent subword tokenizer and detokenizer for neural text processing},</p><p> author={Kudo, Taku and Richardson, John},</p><p> journal={arXiv preprint arXiv:1808.06226},</p><p> year={2018}</p><p>}</p><p>mibench</p><p>@inproceedings{polybench2014,</p><p> title={Understanding polybench/c 3.2 kernels},</p><p> author={Yuki, Tomofumi},</p><p> booktitle={International workshop on polyhedral compilation techniques (IMPACT)},</p><p> pages={1--5},</p><p> year={2014}</p><p>}</p><p>polybench: </p><p>@inproceedings{mibench,</p><p> title={MiBench: A free, commercially representative embedded benchmark suite},</p><p> author={Guthaus, Matthew R and Ringenberg, Jeffrey S and Ernst, Dan and Austin, Todd M and Mudge, Trevor and Brown, Richard B},</p><p> booktitle={4th IEEE international workshop on workload characterization},</p><p> year={2001}</p><p>}</p>
"An efficient ptychography reconstruction strategy through fine-tuning of large pre-trained deep learning model" train and test data
<ul><li>Model for the article "An efficient ptychography reconstruction strategy through fine-tuning of large pre-trained deep learning model".</li><li>The .pth file is the pre-trained PtyNet-S model and the fine-tuned PtyNet-B model.</li><li>Please contact panxy@ihep.ac.cn if you have any questions.</li></ul>
Supplementary material of article "Stem-loop-induced ribosome queuing in the uORF2/ATF4 overlap fine-tunes stress-induced human ATF4 translational control"
Open the record for dataset details and reuse information.
Protein sequences for Dephosphorylation sites and fine-tuning notebook
<p>Protein sequences for Dephosphorylation sites</p>
Black-box Membership Inference Attacks against Fine-tuned Diffusion Models
<p>We have provided some fine-tuned model checkpoints and datasets to help readers reproduce the experiments presented in the paper.</p>
Data Supporting The Paper 'Fine-tuned spatiotemporal dynamics of DNA replication during phage lambda infection'
<p>The dataset includes raw images, source code, raw vectors saved from MATLAB, and curated data used to generate figures and analyses in the paper <strong>'Fine-tuned spatiotemporal dynamics of DNA replication during phage lambda infection'</strong> by Z. Yu, et al.</p> <p>Additional information about the experiments will be available upon request. </p>
Supplementary data for eEDM study in thesis 'Low energy observables and fine-tuning in the MSSM'
<p>For the eEDM study in Chapter 5 of the thesis 'Low energy observables and fine-tuning in the MSSM' we have created several input sets, one for each run. These can be found as 'input_iter[x].csv' and are used to created a SPheno input file. The code to run all the software is found in the directory 'Spheno_to_eEDM'. A description of this code can be found in the thesis (DOI will follow upon succesful defense).</p> <p>For each iteration we have gathered the relevant output (masses, couplings, mixing matrices, observables, fine-tuning etc) from the different output files. These are stored in csv files, gathered in eEDMdata.tar . One line in the output file corresponds to one line in the original input file, where the directory name (dir_name) in the output file is the same as the index of the input file. This is one data point of the study.</p>
No More In-Context Learning? Exploring Parameter-Efficient Fine-Tuning Techniques for Code Generation with Large Language Models
<p>Data and models part of the replication package of the ICSE 24 submission entitled "<em>No More In-Context Learning? Exploring Parameter-Efficient Fine-Tuning Techniques for Code Generation with Large Language Models</em>".</p>
Data and code for ""Smoking gun" signatures of topological milestones in trivial materials by measurement fine-tuning and data postselection"
<p><strong>Brief description</strong></p> <p>This repository contains data, code, and other materials for ""Smoking gun" signatures of topological milestones in trivial materials by measurement fine-tuning and data postselection". Files are zipped by project.</p> <p><strong>Data formats</strong></p> <ul> <li>MTX: A simple 2D/3D matrix format developed for <a href="https://nsweb.tn.tudelft.nl/~gsteele/spyview/">Spyview</a>.</li> <li>DAT: Plain-text tabular data.<br> DAT and MTX files can be plotted with Spyview, qtplot (a portable version for Windows can be downloaded <a href="https://github.com/cover-me/qtplot/releases/download/2020.09.21/qt_plot.2020.09.21.7z">here</a>), Jupyter notebooks in ZIP files or <a href="https://github.com/cover-me/qtview">here</a>.</li> <li>SET: Instrument settings.</li> <li>PY: Measurement scripts.</li> <li>IPYNB or HTML: Jupyter notebooks with code and figures. IPYNB can be previewed on this <a href="https://kokes.github.io/nbviewer.js/viewer.html">page</a>.</li> </ul> <p><strong>Additional repositories</strong></p> <p>MATLAB codes for theoretical simulations in Ref 63: https://github.com/frolovgroup/</p> <p>Full data and code from the Shapiro step project: [https://zenodo.org/record/6416083](Data and code for "smash junction" papers)</p>
Participatory Research for Fine-tuning of a 2.0 System to Optimise Home Management of Oral Cancer Therapies.
ClinicalTrials.gov study NCT02921724. IPD Sharing: UNDECIDED. Countries: 1. Publications: 2.
Data from: Causes and consequences of repeatability, flexibility and individual fine-tuning of migratory timing in pike
Open the record for dataset details and reuse information.
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.