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.
8
datasets available to search
ShareScore release 0.9.0
Dataset results
8 results for “parameter tuning”
Dataset for the fine-tuning of parameters and hyper parameters, and the evaluation of the Scorca agent
<p># Scorca Data<br>This repository contains data collected during multiple tests of the Scorca agent.</p> <p>This data was used to generate the various plots seen in the ICAART paper titled ***Knowledge Modelling, Strategy Designing, and Agent Engineering for Reconnaissance Blind Chess***, and to forge decisions for the agent, i.e. with which strategy to go and which hyperparameters to use. </p> <p>It also contains multiple scripts used to generate the plots in the paper.</p> <p>## Project overview</p> <p>The project is organized into several key directories, each containing specific components of the Scorca agent's testing and data analysis:</p> <p>- /experiments: Contains various experimental data and scripts.<br> - /entropy_comp: Data and scripts related to entropy computation experiments.<br> - /piece_states_removal: Information on experiments involving the removal of piece states.<br> - /sense_comp: Contains sub-directories for adapted entropy, likely senses, and opponent move weight analysis.<br>- /misc_src: Miscellaneous source files and scripts.<br> - /efficiency: Scripts and data related to the efficiency analysis of the agent.<br> - /enemy_bot_movement: Data on enemy bot movement patterns and strategies.<br> - /history: Historical data and analysis scripts.<br> - /naive_entropy: Scripts for naive entropy calculations.<br>- /sense_comp: Sense computation related files.<br> - /all_possible_states: Tracking and analysis of all possible board states.<br> - /likely_senses: Data on the most likely senses used in various game scenarios.<br> - /opp_move_weight: Analysis of opponent move weight in different contexts.</p> <p>For the agent source code, please refer to the Scorca GitHub repository: https://github.com/Robinbux/Scorca.</p> <p>- Contact Robin in case of any inquiry (rb.stoehr@gmail.com)<br>- The licence is CC-BY 4.0.</p>
Supplementary materials to the paper: Automatic Parameters Tuning of Late Reverberation Algorithms for Audio Augmented Reality
<p>Supplementary materials to the paper:</p> <blockquote> <p>Riccardo Bona, Davide Fantini, Giorgio Presti, Marco Tiraboschi, Isaac Engel and Federico Avanzini. 2022. Automatic Parameters Tuning of Late Reverberation Algorithms for Audio Augmented Reality. In <em>Proceedings of International Conference on Audio Mostly</em>.</p> </blockquote> <p>The supplementary materials include the reverberated audio stimuli employed in the MUSHRA listening test reported in the paper. For each type of audio stimuli (Drums, Sax and Speech) the version reverberated with each of the six target Room Impulse Responses (RIRs) is provided along with the versions reverberated using the reverb matching method proposed in the paper (two different artificial reverberators have been considered: FDN and Freeverb).</p> <p>Further, the reverberation times (<span class="math-tex">\(T_{20}\)</span>) per octave band for each considered RIR are provided.</p>
Data for "Tuning parameters of dimensionality reduction methods for single-cell RNA-seq analysis"
<p>The files named <code>df_scran.csv</code>, <code>df_seurat.csv</code>, <code>df_zinbwave.csv</code>, <code>df_dca.csv</code>, and <code>df_scvi.csv</code> contain one row per configuration that we ran successfully.</p> <p>The files named <code>DATASET.METHOD.h5ad</code> are encoded with anndata <code>v0.7.0</code> (be careful as they are not readable with previous versions) and contain 100 embeddings each. The embeddings are in the <code>obsm</code> attribute of the object. All the embeddings can be listed with the <code>obsm_keys()</code> method. The name of the embedding contains the parameters used to generate that embedding and are written like that <code>method=zinbwave.dims=10.epsilon=1000.features=300.gene_covariate=0</code>.</p> <p> </p> <p>For questions on this dataset please contact fraimundo@google.com</p>
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.
Supplementary Dataset for the Paper: "Parameter-Adaptive Approximate MPC: Tuning Neural-Network Controllers without Re-Training"
<p>Supplementary dataset for paper Hose, Henrik, Alexander Gräfe, and Sebastian Trimpe. "Parameter-Adaptive Approximate MPC: Tuning Neural-Network Controllers without Re-Training." <em>arXiv preprint arXiv:2404.05835</em> (2024).</p> <p>The code to use this dataset is publicly available at <a href="https://github.com/hshose/Adaptive-AMPC-Cartpole">https://github.com/hshose/Adaptive-AMPC-Cartpole</a></p> <p>The dataset contains training and testing data to train an NN controller for a standart cartpole system.</p> <p>For each system, there are initial conditions as comma separated value in the `x0.txt` file, the MPC input trajectory in the `U.txt` file and the corresponding predicted state sequence in the `X.txt` file. Additionally, sensitivities (i.e. gradients dU/dtheta with respect to some system parameters theta) are provided in a file called `J.txt`.</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>
Med-ReLU: A Hybrid Activation Function Tailored for Deep Artificial Neural Networks in Medical Image Segmentation without Parameter Tuning
<p>Background: Deep learning (DL) is derived from the domain of Artificial Neural Network (ANN). It makes one of the most important elements of deep learning algorithms. Deep learning segmentation models are based on layer-by-layer convolution learning attribute representation directed by forward and backward propagation. Throughout the process vital role is played by appropriately chosen activation function (AF) in order to guarantee the robustness of the model learning. However, the existing activation functions are either ineffective in addressing the vanishing gradient problem or get burdened with multiple parameters that need to be manually tuned. Moreover, the current research on activation function design mainly focuses on classification tasks using natural images from the MNIST, CIFAR-10 and CIFAR-100 datasets. Therefore,Med-ReLU as a novel activation function for medical image segmentation, is proposed. The proposed activation function avoids deep learning models from the attacks of dead neurons or from the vanishing gradient problems. Method: Med-ReLU is a hybrid activation function that combines the property of two activation functions of ReLU and Softsign. For positive inputs, Med-ReLU utilizes the linear property just like ReLU to produce an output without vanishing gradient. The negative inputs converge in polynomial ways towards their asymptotes as property of the softsign AF that ensures robust training processing without the problem of dead neurons that rarely activate across the entire training dataset. Results: The training performance and segmentation accuracy of Med-ReLU have been investigated. The proposed function has demonstrated stable training and does not suffer from over-fitting. Hence, Med-ReLU has consistently outperformed the existing state-of-art activation functions in medical image segmentation tasks. Conclusion: Med-ReLU has been designed as a parameter-free activation function for DL image segmentation tasks. This activation function is easy-to-implement on complex and deep learning models. The utility of this research lies in affirming the impact of Med-ReLU on different Artificial Neural Network architectures and for various kinds of anomaly addressing tasks.</p>
Data for the bachelor thesis "Investigating Tuning Parameters in ECHAM-HAM with a Perturbed Parameter Ensemble"
<p>This repository contains the data for the bachelor thesis:<br> <br> Author: Melina Abeling<br> Title: Investigating Tuning Parameters in ECHAM-HAM with a Perturbed Parameter Ensemble<br> Date: July 2022</p> <p><br> The scripts are to be found in the accompanying package (https://doi.org/10.5281/zenodo.6866140).</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.