Find research datasets worth reusing

opencc-by-4.0Dec 2017View details →

Visualizing histopathologic deep learning classification and anomaly detection using nonlinear feature space dimensionality reduction

Representative Testing/Validation WSIs used in the manuscript "Visualizing histopathologic deep learning classification and anomaly detection using nonlinear feature space dimensionality reduction"

opencc-by-4.0Dec 2017View details →

Visualizing histopathologic deep learning classification and anomaly detection using nonlinear feature space dimensionality reduction

Training image dataset used in the manuscript "Visualizing histopathologic deep learning classification and anomaly detection using nonlinear feature space dimensionality reduction"

opencc-by-4.0Feb 2019View details →

AbDb processed and pickled for use in deep learning CDR-H3 Structure prediction

This is a pickle file, ready for training by the neural network described in "Improving CDR-H3 modelling in Antibodies" found at the following URL: <a href="https://github.com/OniDaito/MRes">https://github.com/OniDaito/MRes</a> The data is derived from the AbDb dataset found at: <a href="http://www.bioinf.org.uk/abs/abdb/">http://www.bioinf.org.uk/abs/abdb/</a>

opencc-by-4.0May 2019View details →

Data for "On the impact of Citizen Science-derived data quality on deep learning based classification in marine images"

This dataset contains all the annotations done by either citizen scientists or experts of the publication: "On the impact of Citizen Science-derived data quality on deep learning based classification in marine images" CSP.csv -> CS annotations of the Citizen Science Primer-experiment CSPExpert.csv -> Expert annotations of the Citizen Science Primer-experiment CSS.csv -> CS annotations of the Citizen Science Study CSSExpert.csv -> Expert annotations of the Citizen Science Study Visual exploration of the image data is possible in the BIIGLE 2.0 image annotation system at <a href="https://biigle.de/projects/139">https://biigle.de/projects/159</a> using the login <a href="mailto:cs@example.com">cs@example.com</a> and the password plosonecs.

opencc-by-4.0Jul 2019View details →

SLDeep: Statement-Level Software Defect Prediction Using Deep-Learning Models on Static Code Features

Software defect prediction (SDP) seeks to estimate fault-prone areas of the code to focus testing activities on more suspicious portions. Consequently, high-quality software is released with less time and effort. The current SDP techniques however work at coarse-grained units, such as a module or a class, putting some burden on the developers to locate the fault. To address this issue, we propose Statement-Level software defect prediction using Deep-learning model (SLDeep). To reify our proposal, we defined a suite of 32 statement-level metrics, such as the number of binary and unary operators used in a statement. Then, we applied as learning model, long short-term memory (LSTM). The significance of SLDeep for intelligent and expert systems is that it demonstrates a novel use of deep-learning models to the solution of a practical problem faced by software developers. We conducted experiments using more than 100,000 C/C++ programs within the Code4Bench. The programs total 2,356,458 lines of code with 292,064 faulty lines. The benchmark comprises diverse set of programs and versions, written by thousands of developers. Therefore, it tends to give a model that can be used for cross-project SDP. In the experiments, our trained model could successfully classify the unseen data with average performance measures 0.945, 0.971, and 0.976 in terms of recall, precision, and accuracy, respectively. These experimental results suggest that SLDeep is effective for statement-level SDP. The impact of this work is twofold. Working at statement-level further alleviates developer’s burden in pinpointing the fault locations. Second, cross-project feature of SLDeep helps defect prediction research become more industrially-viable for more information visit <a href="https://github.com/sldeep/SLDeep">https://github.com/sldeep/SLDeep</a>

opencc-by-4.0Jun 2019View details →

Industry-scale Application and Evaluation of Deep Learning for Drug Target Prediction

Artificial intelligence (AI) is undergoing a revolution thanks to the breakthroughs of machine learning algorithms in computer vision, speech recognition, natural language processing and generative modelling. Recent works on publicly available pharmaceutical data showed that AI methods are highly promising for Drug Target prediction. However, the quality of public data might be different than that of industry data due to different labs reporting measurements, different measurement techniques, fewer samples and less diverse and specialized assays. As part of a European funded project (ExCAPE), that brought together expertise from pharmaceutical industry, machine learning, and high-performance computing, we investigated how well machine learning models obtained from public data can be transferred to internal pharmaceutical industry data. Our results show that machine learning models trained on public data can indeed maintain their predictive power to a large degree when applied to industry data. Moreover, we observed that deep learning derived machine learning models outperformed comparable models, which were trained by other machine learning algorithms, when applied to internal pharmaceutical company datasets. To our knowledge, this is the first large-scale study evaluating the potential of machine learning and especially deep learning directly at the level of industry-scale settings and moreover investigating the transferability of publicly learned target prediction models towards industrial bioactivity prediction pipelines.

opencc-zeroJul 2024View details →

Deep Learning Neural Network Development for the Classification of Bacteriocin Sequences Produced by Lactic Acid Bacteria

This project contains the following underlying data: <h3>      Software-Related Files</h3>                ·         BacLABNet_script.ipynb (Deep Learning Neural Network for classification of Bacteriocin Sequences)                 ·         embed_proteins.py (Recurrent Neural Network to obtained the embedding vectors)                ·         model_I22.h5 (This file contains the trained weights of the trained model)                ·         model_I22.json (This file contains the structure of the trained model)                ·         rnn_gru.pt (Initial weights of the Recurrent Neural Network to obtain embedding vectors)                ·         List_kmers.csv (List of 5-mers and 7-mers obtained from dataset after it filtered sequences shorter than 50 aa and                           longer than 2000 aa)                     Files Used for Training, Testing, and Validation of the Neural Network                ·         data_nonBacLAB.csv (25000 nonBacLAB amino acid sequences retrieved from Uniprot)                ·         data_BacLAB.csv (24964 BacLAB amino acid sequences retrieved from Uniprot)              Additional Files                 ·         data_BacLAB_and_nonBacLAB.csv (Combination of sequences from data_BacLAB.csv and data_nonBacLAB.csv)                  ·         all k.mers list.xlsx (Table of all k-mers obtained for k=3,5,7,15,20)   Note: Codes are additionally available on GitHub.  Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication)(http://creativecommons.org/publicdomain/zero/1.0/)

opencc-by-4.0Oct 2023View details →

UniSpec: Deep Learning for Predicting the Full Range of Peptide Fragment Ion Series to Enhance the Proteomics Data Analysis Workflow

UniSpec is a comprehensive DL spectrum predictor that can predict the intensity of the entire HCD MS/MS fragment ion series, going beyond existing tools limited to b/y ion series.  All datasets developed for UniSpec model are shared on Zenodo as part of the UniSpec publication, "UniSpec: Deep Learning for Predicting Comprehensive Peptide Fragment Ion Series to Improve Peptide-Spectrum Matches from Shotgun Proteomics Experiments". This includes UniSpec datasets, downstream evaluation and analysis, and application case studies. 1. pre-processed training, evaluation and testing data for machine learning;          UniSpec-Datasets.7z, Readme_UniSpecDatasets.txt 2. Streamlined  input datasets based on the fragmentation dictionary;         Streamlined_inputdatasets.7z, Readme_Streamlined_inputdatasets.txt 3. Predictions on the validation and test sets;        UniSpecPred_Validation-Test.7z, Readme_Predictons_ValidationTest.txt 4. Evaluation by comparison with Prosit;       a. Predictions: prosit_and_unispec_predictions.7z, Readme_prosit_and_unispec_predictions.txt       b. Cosine similarity scores: prosit_vs_unispec_CS.7z, Readme_prosit_vs_unispec_CS.txt 5. CSS for Different HCD Fragment Ion Series;        CS_for_ion_splits.tsv 6. Application 1: PSM rescoring;       PSM rescoring_zipfiles.7z,  PSM rescoring_readme.txt 7. Application 2: In-silico spectral library search         in-silico_librarysearch.7z, in-silico_librarysearch_readme.txt

opencc-by-4.0Mar 2024View details →

Exploring deep learning models for 4D-STEM-DPC data processing

This repository contains scanning transmission electron microscopy data and processing files used in the journal publication "Exploring deep learning models for 4D-STEM-DPC data processing". DOI: <a href="https://doi.org/10.1016/j.ultramic.2024.114058">10.1016/j.ultramic.2024.114058</a> Prerequisites The scripts presented below require certain open-source Python packages to run. Library versions used to run the scripts are: <ul> <li>hyperspy 1.7.1</li> <li>pyxem 0.14.2</li> <li>fpd 0.2.5</li> <li>pytorch 1.12.1 (cudatoolkit 11.6.0)</li> <li>jupyterlab 4.0.7</li> </ul> Data files Three zipped folders are included. Two of them contain the training- and inference data for the neural networks, aptly named training_data.zip and inference_data.zip. PyTorch state dictionaries for trained models are included in the models.zip folder. Processing scripts All scripts are included in an IPython notebook format (.ipynb extension). The notebooks Segmentation.ipynb and Regression.ipynb contain the code for training and inference of the segmentation and regression models, respectively. The Training_data_creation.ipynb notebook contains the code to preprocess the training data for both neural network models. The Standard_algorithms.ipynb notebook has the code for doing center of mass and edge filtering/disc detection algorithms for STEM-DPC processing.

Replication Package for "PyTraceBERT: Python Traceback-based Language Model for Detecting Compatibility Issues in Deep Learning Systems"

This package contains the traceback data, pre-trained models, and static word embeddings used in the paper, PyTraceBERT: Python Traceback-based Language Model for Detecting Compatibility Issues in Deep Learning Systems.

Detection of Areas with Human Vulnerability Using Public Satellite Images and Deep Learning (Dataset)

<div> <h2>Overview</h2> <a href="https://github.com/fbvidal/HumanVulnerabilityDetectionDL#overview"></a></div> This repository contains the code and resources for the project titled "Detection of Areas with Human Vulnerability Using Public Satellite Images and Deep Learning". The goal of this project is to identify regions where individuals are living under precarious conditions and facing neglected basic needs, a situation often seen in Brazil. This concept is referred to as "human vulnerability" and is exemplified by families living in inadequate shelters or on the streets in both urban and rural areas. Focusing on the Federal District of Brazil as the research area, this project aims to develop two novel public datasets consisting of satellite images. The datasets contain imagery captured at 50m and 100m scales, covering regions of human vulnerability, traditional areas, and improperly disposed waste sites. The project also leverages these datasets for training deep learning models, including YOLOv7 and other state-of-the-art models, to perform image segmentation. A comparative analysis is conducted between the models using two training strategies: training from scratch with random weight initialization and fine-tuning using pre-trained weights through transfer learning. <div> <h3>Key Achievements</h3> <a href="https://github.com/fbvidal/HumanVulnerabilityDetectionDL#key-achievements"></a></div> <ul> <li>Two new satellite image datasets focusing on human vulnerability and improperly disposed waste sites, available in public domains.</li> <li>Comparison of image segmentation models, including YOLOv7 and Segmentation Models, with performance metrics.</li> <li>Best F1-scores: 0.55 for YOLOv7 and 0.64 for Segmentation Models.</li> </ul> This repository provides the code, models, and data pipelines used for training, evaluation, and performance comparison of these deep learning models. <div> <h2>Citation (Bibtex)</h2> <a href="https://github.com/fbvidal/HumanVulnerabilityDetectionDL?tab=readme-ov-file#citation-bibtex"></a></div> <pre><code>@TECHREPORT {TechReport-Julia-Laura-HumanVulnerability-2024, author = "Julia Passos Pontes, Laura Maciel Neves Franco, Flavio De Barros Vidal", title = "Detecção de Áreas com Atividades de Vulnerabilidade Humana utilizando Imagens Públicas de Satélites e Aprendizagem Profunda", institution = "University of Brasilia", year = "2024", type = "Undergraduate Thesis", address = "Computer Science Department - University of Brasilia - Asa Norte - Brasilia - DF, Brazil", month = "aug", note = "People living in precarious conditions and with their basic needs neglected is an unfortunate reality in Brazil. This scenario will be approached in this work according to the concept of \"human vulnerability\" and can be exemplified through families who live in inadequate shelters, without basic structures and on the streets of urban or rural centers. Therefore, assuming the Federal District as the research scope, this project proposes to develop two new databases to be made available publicly, considering the map scales of 50m and 100m, and composed by satellite images of human vulnerability areas, regions treated as traditional and waste disposed inadequately. Furthermore, using these image bases, trainings were done with the YOLOv7 model and other deep learning models for image segmentation. By adopting an exploratory approach, this work compares the results of different image segmentation models and training strategies, using random weight initialization (from scratch) and pre-trained weights (transfer learning). Thus, the present work was able to reach maximum F1 score values of 0.55 for YOLOv7 and 0.64 for other segmentation models." } </code></pre> <div> </div> <div> <h2>License</h2> <a href="https://github.com/fbvidal/HumanVulnerabilityDetectionDL?tab=readme-ov-file#license"></a></div> This project is licensed under the MIT License - see the LICENSE file for details.

Winter Precipitation-Type Models for "Evidential Deep Learning: Enhancing Predictive Uncertainty Estimation for Earth System Science Applications"

This contains trained model weights, scalers, and evaluation metrics for the winter precipitation-type models trained as part of the paper "Evidential Deep Learning: Enhancing Predictive Uncertainty Estimation for Earth System Science Applications".

Supplementary Data: Data Imbalance in Drug Response Prediction: Multi-Objective Optimization Approach in Deep Learning Setting

Open the record for dataset details and reuse information.

UML Diagram Dataset from the paper Creating and Validating a Ground Truth Dataset of UML Diagrams Using Deep Learning Techniques

Dataset of six UML diagram classes, comprising a total of 2,626 images (426 activity diagrams, 636 class diagrams, 352 component diagrams, 357 deployment diagrams, 435 sequence diagrams, and 420 use case diagrams). Importantly, unlike other existing datasets, ours contains no duplicate elements and all diagrams are correctly classified.

MultiFranceFences: A novel deep learning dataset for automated fence detection from multimodal aerial imagery

The MultiFranceFences dataset is a large-scale, multimodal remote sensing benchmark for the semantic segmentation of fences across various landscapes in France. This dataset integrates high-resolution orthophotographs (RGB through BDOrtho) and Digital Surface Models (DSM) derived from LiDARHD data.  MultiFranceFences is suitable for deep learning models in semantic segmentation, including state-of-the-art models like UNet, D-LinkNet, and the newly proposed H-IncepUNet, which integrates handcrafted features and multi-scale feature extraction modules for enhanced fence detection. Dataset features: <ul> <li>Multimodal imagery: Combines orthophotographs and DSM data from LiDARHD for fences semantic segmentation (folders ortho and lidar).</li> <li>Buffer options: 2-meter and 3-meter buffer fence annotations to fit varying detection requirements (folders fences_2m and fences_3m).</li> <li>Diverse landscapes: Covers rural, and natural environments across France.</li> <li>Validated dataset: Manually cleaned and validated to remove erroneous fence labels under tree canopies or areas with limited visibility.</li> </ul> Each patch is named according to the nomenclature of the original BDOrtho tile, followed by the specific x and y coordinates of the patch within that tile.

Companion data artifacts: Technical framework demonstration for deep learning-based wood species classification with advanced sub-μ-CT imaging

This is the companion data artifact collection for the IWAWA paper manuscript by Jannik Stebani, Tim Lewandrowski, Kilian Dremel, Simon Zabler and Volker Haag.It is generally to be used with the visualization and prediction showcases implemented in the Binder notebooks launched from this <a href="https://github.com/stebix/woodnet-showcase" target="_blank" rel="noopener">woodnet-showcase</a> GitHub repository. The artifacts amount to the following: <ol> <li><code>acer-artifact.hdf5</code> : Exemplary <code>(256, 256, 256)</code> subvolume from a Acer pseudoplatanus sub-μ-CT scan</li> <li><code>pinus-artifact.hdf5</code> : Exemplary <code>(256, 256, 256)</code> subvolume from a Pinus sylvestris sub-μ-CT scan</li> <li><code>weights-artifact.pth</code> : Exemplary PyTorch trained weights for a woodnet/deep neural network to demonstrate classification of the above samples</li> </ol>

Global perspectives on sand dune patterns: Scale-adaptable classification using Landsat imagery and deep learning strategies

Here we generated the global sand dune pattern map at a resolution of 30 m, named GSDP30. The GSDP30 map encompasses 11 types of sand dune patterns (SDPs): simple crescentic dunes, compound-complex crescentic dunes, simple linear dunes, compound-complex linear dunes, dome dunes, star dunes, parabolic dunes, dendritic dunes, network dunes, sand sheets, and others. The map is divided into 331 Tiff tiles, each characterized by a size of 15,360 × 15,360 pixels and named according to the longitude and latitude coordinates of its upper-left corner.

openmit-licenseOct 2024View details →

Supporting data for RCANE: A Deep Learning Algorithm for Whole-genome Pan-Cancer Somatic Copy Number Aberration Prediction using RNA-seq Data.

This is the data repository for RCANE: A Deep Learning Algorithm for Whole-genome Pan-Cancer Somatic Copy Number Aberration Prediction using RNA-seq Data. To use this dataset, please refer to <a href="https://github.com/HowardGech/RCANE" target="_blank" rel="noopener">https://github.com/HowardGech/RCANE</a>.

opencc-by-4.0Mar 2023View details →

A Submesoscale Eddy Identification Dataset Derived from GOCI I Chlorophyll–a Data based on Deep Learning

This is an observational dataset on submesoscale eddies, which obtains from high–resolution chlorophyll–a distribution images from GOCI I. We employed a combination of digital image processing, filtering, YOLOv7–X, and small object detection techniques, along with specific chlorophyll image enhancement processing, to extract information on submesoscale eddies, including their time, polarity, geographical coordinates of the eddy center, eddy radius, coordinates of the upper left and lower right corners of the prediction box, area of the eddy's inner ellipse, and confidence score, which covers eight daily periods between 00:00 and 08:00 (UTC) from April 1, 2011, to March 31, 2021. We identified a total of 19,136 anticyclonic eddies and 93,897 cyclonic eddies at a confidence threshold of 0.2. The mean radius of anticyclonic eddies is 24.44 km (range 2.5 km to 44.25 km), while that of cyclonic eddies is 12.34 km (range 1.75 km to 44 km). The unprecedented hourly resolution dataset on submesoscale eddies provides information on their distribution, morphology, and energy dissipation, making it a significant contribution to understanding marine environments and ecosystems, as well as improving climate model predictions. The article doi associated with this dataset is https://doi.org/10.5194/essd-2024-188.

Enhancing Smartphone Battery Life: A Deep Learning Model Based on User-Specific Application and Network Behaviour

This work presents an analysis based on training AI models directly on devices to make personalized predictions tailored to individual usage patterns, ensuring that each user benefits from a personalized approach to battery management. By integrating these AI-based insights, mobile devices can proactively manage power consumption, improving battery performance and user satisfaction. This personalized, intelligent approach to battery management represents a significant advance in optimizing device efficiency and addresses the growing demand for longer-lasting mobile technology.