Share your research data

The high-throughput dynamic usage dataset developed in this study consists of 81 commercial high-nickel NMC811/Gr-SiOx cells (LG Chem, INR18650-MJ1, nominal capacity of 3.5 Ah and nominal voltage of 3.635 V). All experiments were performed at a controlled temperature of 25 °C. The relaxation dataset contains voltage responses at different SOCs during rest steps, obtained from RPTs under each usage scenario. The periodic RPTs were carried out every 100 cycles using the diagnostic protocol, which sampled at 2.5% SOC intervals within the 40–80% SOC range. At each SOC point, two sequences were applied: (i) 0.1C charging, a 10 min Rest1, a 2C 10 s pulse, followed by a 20 min Rest2; and (ii) 0.5C charging, a 30 min Rest1, a 2C 10 s pulse, followed by a 20 min Rest2. The retention dataset contains capacity retention data for all cells across both cycling and RPTs.

Experimental and simulated light-off curves

Código en lenguje python para realizar la evaluación de la viabilidad económica de cada uno de los puntos estudiados de con potencial de recuperación de energía alto, obtenidos mediante la jerarquización de las tuberías de las red de distribución de agua del municipio de Cartago, Valle del Cauca, Colombia, en el estudio "RECUPERACIÓN DE ENERGÍA EN UNA RED MATRIZ DE UN SISTEMA DE ABASTECIMIENTO DE AGUA POTABLE "

Raw dataset for all figures in the manuscript by Siddhant et al., to be published as, "Sediment-water interface dissolution rate as key control on depth-dependent deep-sea calcite diagenesis: insights from Walvis Ridge cores (IODP 391)".

This is the code that we have used for the " Influence of Social Belonging on Adolescents' Internet and Phone Dependency" paper

this dataset contains the information of A-share listed companies in China from 2014 to 2020. In order to ensure the reliability, the sample is processed as follows: (1) excluding companies in the financial and insurance industries; (2) excluding ST, ST*, PT, and insolvent companies; (3) excluding companies with missing data on major variables; and (4) in order to avoid the effect of extreme values, all continuous variables are tested at the 1% and 99% quartiles for shrinkage treatment. After the treatment, a total of 11,051 valid observations from 1798 listed enterprises were obtained. The green innovation data of enterprises are from CNRDS database, and the basic information and financial data of enterprises are from CSMAR database.

This dataset contains 36 synthetic fruit images generated using the Python PIL library. It includes three categories of fruits: Apple, Banana, and Orange, with 12 images per class. Each image has a resolution of 224×224 pixels in RGB PNG format and is properly labeled. The dataset is primarily designed for educational and research purposes, including: - Multi-class image classification tasks - Introductory computer vision practice - Demonstration of dataset creation and publishing on Mendeley Data File Structure: ├── apple/ → 12 images ├── banana/ → 12 images └── orange/ → 12 images Key Features: - 3 fruit categories (apple, banana, orange) - 36 images in total - 224×224 pixels, RGB, PNG format - Synthetic illustrations (not real photographs) - Suitable for classification tasks, teaching, and dataset publishing demonstrations ... License: CC BY 4.0 Keywords: Fruits, Image Classification, Computer Vision, Synthetic Dataset, Machine Learning

This collection brings together six datasets designed for experiments in text segmentation: Choi — 922 artificial documents [1]. Each document is composed of sentence blocks drawn from different sources. Since the segments are unrelated, segmentation is relatively easy for many algorithms and typically yields high accuracy. Manifesto — 6 long political speeches [2]. Each text includes a human-generated segmentation based on strict guidelines. The dataset is used to evaluate segmentation of semantic topic shifts and thematic changes. Wiki-1024 — 1,024 Wikipedia articles [3]. Segmentation is defined by the natural division of documents into sections and subsections. Abstracts — artificial documents created by merging real research abstracts into continuous texts. About 20,000 abstracts were collected from Scopus in the field of Information Retrieval. Segments correspond directly to individual abstracts. SMan — artificial documents constructed by randomly sampling segments from Manifesto texts. The resulting statements vary in content due to mixing, but generally follow a slogan-like style. PhilPapersAI — 336 philosophy articles (focused on AI) selected from philpapers.org. The source PDFs were reprocessed using the OpenAI GPT-4o-mini LLM to restore structure and add subsection divisions. The resulting texts are coherent and well-structured, while preserving the authors’ original style as closely as possible. [1] Choi, F.Y.Y. Advances in domain independent linear text segmentation. In Proceedings of the 1st Meeting of the North American Chapter of the Association for Computational Linguistics, 2000. [2] Hearst, M.A. Text Tiling: Segmenting Text into Multi-paragraph Subtopic Passages. Computational Linguistics 1997, 23, 33–64. [3] Koshorek, O.; Cohen, A.; Mor, N.; Rotman, M.; Berant, J. Text Segmentation as a Supervised Learning Task. In Proceedings of the Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers); Walker, M.; Ji, H.; Stent, A., Eds., New Orleans, Louisiana, 2018; pp. 469–473. https://doi.org/10.18653/v1/N18-2075.

outcomes

for Gut Microbes