Trending Public Datasets Overview

These Datasets contain a diverse collection of datasets intended for machine learning research and practice. Each dataset is curated to support different types of machine learning challenges, including classification, regression, and clustering. Below is a detailed list of the datasets available in this repository, along with descriptions and links to their sources.

Available Datasets

Iris Dataset

Description: This classic dataset includes measurements for 150 iris flowers from three different species. It includes four features: sepal length, sepal width, petal length, and petal width. Source: Iris Dataset Source Files: iris.csv

DHFR Dataset

Description: Contains data for 325 molecules with biological activity against the DHFR enzyme, relevant in anti-malarial drug research. It includes 228 molecular descriptors as features. Source: DHFR Dataset Source Files: dhfr.csv

Heart Disease Dataset (Cleveland)

Description: Comprises diagnostic measurements from 303 patients tested for heart disease at the Cleveland Clinic. It features 13 clinical attributes. Source: UCI Machine Learning Repository Files: heart-disease-cleveland.csv

HCV Data

Description: Detailed datasets related to Hepatitis C Virus (HCV) progression, with features for classification and regression tasks. Files: HCV_NS5B_Curated.csv, hcv_classification.csv, hcv_regression.arff

NBA Seasons Stats

Description: Player statistics from the NBA 2020 and 2021 seasons for detailed sports analytics. Files: NBA_2020.csv, NBA_2021.csv

Boston Housing Dataset

Description: Data concerning housing values in the suburbs of Boston, suitable for regression analysis. Files: BostonHousing.csv, BostonHousing_train.csv, BostonHousing_test.csv

Acetylcholinesterase Inhibitor Bioactivity

Description: Chemical bioactivity data against acetylcholinesterase, a target relevant to Alzheimer's research. It includes raw and processed formats with chemical fingerprints. Files: acetylcholinesterase_01_bioactivity_data_raw.csv to acetylcholinesterase_07_bioactivity_data_2class_pIC50_pubchem_fp.csv

California Housing Dataset

Description: Data aimed at predicting median house prices in California districts. Files: california_housing_train.csv, california_housing_test.csv

Virtual Reality Experiences Data

Description: Data from user experiences with various virtual reality setups to study user engagement and satisfaction. Files: Virtual Reality Experiences-data.csv

Fast-Food Chains in USA

Description: Overview of various fast-food chains operating in the USA, their locations, and popularity. Files: Fast-Food Chains in USA.csv

Contributing We welcome contributions to this dataset repository. If you have a dataset that you believe would be beneficial for the machine learning community, please see our contribution guidelines in CONTRIBUTING.md.

License This dataset is available under the MIT License.

SYNERGY is a free and open dataset on study selection in systematic reviews, comprising 169,288 academic works from 26 systematic reviews. Only 2,834 (1.67%) of the academic works in the binary classified dataset are included in the systematic reviews. This makes the SYNERGY dataset a unique dataset for the development of information retrieval algorithms, especially for sparse labels. Due to the many available variables available per record (i.e. titles, abstracts, authors, references, topics), this dataset is useful for researchers in NLP, machine learning, network analysis, and more. In total, the dataset contains 82,668,134 trainable data points. The easiest way to get the SYNERGY dataset is via the synergy-dataset Python package. See https://github.com/asreview/synergy-dataset for all information.

Machine learning (ML) has gained much attention and has been incorporated into our daily lives. While there are numerous publicly available ML projects on open source platforms such as GitHub, there have been limited attempts in filtering those projects to curate ML projects of high quality. The limited availability of such high-quality dataset poses an obstacle to understanding ML projects. To help clear this obstacle, we present NICHE, a manually labelled dataset consisting of 572 ML projects. Based on evidences of good software engineering practices, we label 441 of these projects as engineered and 131 as non-engineered. In this repository we provide "NICHE.csv" file that contains the list of the project names along with their labels, descriptive information for every dimension, and several basic statistics, such as the number of stars and commits. This dataset can help researchers understand the practices that are followed in high-quality ML projects. It can also be used as a benchmark for classifiers designed to identify engineered ML projects.

GitHub page: https://github.com/soarsmu/NICHE

They are available at https://github.com/nerdyqx/ML. (ZIP)

Context

Labeled datasets are useful in machine learning research.

Content

This public dataset contains approximately 9 million URLs and metadata for images that have been annotated with labels spanning more than 6,000 categories.

Tables: 1) annotations_bbox 2) dict 3) images 4) labels

Update Frequency: Quarterly

Querying BigQuery Tables

Fork this kernel to get started.

Acknowledgements

https://bigquery.cloud.google.com/dataset/bigquery-public-data:open_images

https://cloud.google.com/bigquery/public-data/openimages

APA-style citation: Google Research (2016). The Open Images dataset [Image urls and labels]. Available from github: https://github.com/openimages/dataset.

Use: The annotations are licensed by Google Inc. under CC BY 4.0 license.

The images referenced in the dataset are listed as having a CC BY 2.0 license. Note: while we tried to identify images that are licensed under a Creative Commons Attribution license, we make no representations or warranties regarding the license status of each image and you should verify the license for each image yourself.

Banner Photo by Mattias Diesel from Unsplash.

Inspiration

Which labels are in the dataset? Which labels have "bus" in their display names? How many images of a trolleybus are in the dataset? What are some landing pages of images with a trolleybus? Which images with cherries are in the training set?

This child item describes a machine learning model that was developed to estimate public-supply water use by water service area (WSA) boundary and 12-digit hydrologic unit code (HUC12) for the conterminous United States. This model was used to develop an annual and monthly reanalysis of public supply water use for the period 2000-2020. This data release contains model input feature datasets, python codes used to develop and train the water use machine learning model, and output water use predictions by HUC12 and WSA. Public supply water use estimates and statistics files for HUC12s are available on this child item landing page. Public supply water use estimates and statistics for WSAs are available in public_water_use_model.zip. This page includes the following files: PS_HUC12_Tot_2000_2020.csv - a csv file with estimated monthly public supply total water use from 2000-2020 by HUC12, in million gallons per day PS_HUC12_GW_2000_2020.csv - a csv file with estimated monthly public supply groundwater use for 2000-2020 by HUC12, in million gallons per day PS_HUC12_SW_2000_2020.csv - a csv file with estimated monthly public supply surface water use for 2000-2020 by HUC12, in million gallons per day Note: 1) Groundwater and surface water fractions were determined using source counts as described in the 'R code that determines groundwater and surface water source fractions for public-supply water service areas, counties, and 12-digit hydrologic units' child item. 2) Some HUC12s have estimated water use of zero because no public-supply water service areas were modeled within the HUC. STAT_PS_HUC12_Tot_2000_2020.csv - a csv file with statistics by HUC12 for the estimated monthly public supply total water use from 2000-2020 STAT_PS_HUC12_GW_2000_2020.csv - a csv file with statistics by HUC12 for the estimated monthly public supply groundwater use for 2000-2020 STAT_PS_HUC12_SW_2000_2020.csv - a csv file with statistics by HUC12 for the estimated monthly public supply surface water use for 2000-2020 public_water_use_model.zip - a zip file containing input datasets, scripts, and output datasets for the public supply water use machine learning model version_history_MLmodel.txt - a txt file describing changes in this version

This is a traffic dataset which contains balance size of encrypted malicious and legitimate traffic for encrypted malicious traffic detection. The dataset is a secondary csv feature data which is composed of five public traffic datasets. Our dataset is composed based on three criteria: The first criterion is to combine widely considered public datasets which contain both encrypted malicious and legitimate traffic in existing works, such as the Malwares Capture Facility Project dataset and the CICIDS-2017 dataset. The second criterion is to ensure the data balance, i.e., balance of malicious and legitimate network traffic and similar size of network traffic contributed by each individual dataset. Thus, approximate proportions of malicious and legitimate traffic from each selected public dataset are extracted by using random sampling. We also ensured that there will be no traffic size from one selected public dataset that is much larger than other selected public datasets. The third criterion is that our dataset includes both conventional devices' and IoT devices' encrypted malicious and legitimate traffic, as these devices are increasingly being deployed and are working in the same environments such as offices, homes, and other smart city settings.

Based on the criteria, 5 public datasets are selected. After data pre-processing, details of each selected public dataset and the final composed dataset are shown in “Dataset Statistic Analysis Document”. The document summarized the malicious and legitimate traffic size we selected from each selected public dataset, proportions of selected traffic size from each selected public dataset with respect to the total traffic size of the composed dataset (% w.r.t the composed dataset), proportions of selected encrypted traffic size from each selected public dataset (% of selected public dataset), and total traffic size of the composed dataset. From the table, we are able to observe that each public dataset equally contributes to approximately 20% of the composed dataset, except for CICDS-2012 (due to its limited number of encrypted malicious traffic). This achieves a balance across individual datasets and reduces bias towards traffic belonging to any dataset during learning. We can also observe that the size of malicious and legitimate traffic are almost the same, thus achieving class balance. The datasets now made available were prepared aiming at encrypted malicious traffic detection. Since the dataset is used for machine learning model training, a sample of train and test sets are also provided. The train and test datasets are separated based on 1:4 and stratification is applied during data split. Such datasets can be used directly for machine or deep learning model training based on selected features.

This child item describes a public-supply delivery machine learning model that was developed to estimate public-supply deliveries. Publicly supplied water may be delivered to domestic users or to commercial, industrial, institutional, and irrigation (CII) users. This model predicts total, domestic, and CII per capita rates for public-supply water service areas within the conterminous United States for 2009-2020. This child item contains model input datasets, code used to build the delivery machine learning model, and national predictions. This dataset is part of a larger data release using machine learning to predict public-supply water use for 12-digit hydrologic units from 2000-2020. This page includes the following file: delivery_water_use_model.zip - a zip file containing input datasets, scripts, and output datasets for the delivery water use machine learning model

288,776 anonymized adult chest x-ray dataset in 1024 x 1024 pixel DICOM format with corresponding anonymized free-text reports from Dunedin Hospital, New Zealand between 2010 - 2020. Corresponding radiology reports generated by FRANZCR radiologists were manually annotated for 45 common radiological findings mapped to Unified Medical Language System (UMLS) ontology. Each of the multiclassification annotations contains 4 types of labels, namely positive, uncertain, negative and not mentioned. 33,486 studies were manually labeled. 255,290 were labeled by deep learning models. Accuracy of the AI labeled portion of the dataset with respect to each label will be outlined in the published paper. In the provided dataset, image filenames contain patient index (enabling analysis requiring grouping of images by patients), as well as anonymized date of acquisition information where the temporal relationship between images is preserved. This dataset can be used for training and testing for deep learning algorithms for adult chest x rays.Unfortunately, since Feb 2024, the New Zealand government is changing the data governance on datasets used for AI development and this affects the process of how the CANDID III dataset is to be accessed by the external users. Therefore, the CANDID III dataset is not available for access by users outside Health New Zealand. Further notice of access will be updated here should access by external users be reopened.

Context

There's a story behind every dataset and here's your opportunity to share yours. To help autoML huge test, all datasets contains one target named target

Content

What's inside is more than just rows and columns. Make it easy for others to get started by describing how you acquired the data and what time period it represents, too.

Acknowledgements

We wouldn't be here without the help of others. If you owe any attributions or thanks, include them here along with any citations of past research.

Inspiration

Your data will be in front of the world's largest data science community. What questions do you want to see answered?

This child item describes R code used to determine public supply consumptive use estimates. Consumptive use was estimated by scaling an assumed fraction of deliveries used for outdoor irrigation by spatially explicit estimates of evaporative demand using estimated domestic and commercial, industrial, and institutional deliveries from the public supply delivery machine learning model child item. This method scales public supply water service area outdoor water use by the relationship between service area gross reference evapotranspiration provided by GridMET and annual continental U.S. (CONUS) growing season maximum evapotranspiration. This relationship to climate at the CONUS scale could result in over- or under-estimation of consumptive use at public supply service areas where local variations differ from national variations in climate. This method also assumes that 50% of deliveries for total domestic and commercial, industrial, and institutional deliveries is used for outdoor purposes. This dataset is part of a larger data release using machine learning to predict public supply water use for 12-digit hydrologic units from 2000-2020. This page includes the following file: PS_ConsumptiveUse.zip - a zip file containing input datasets, scripts, and output datasets

From wiki

Kaggle, a subsidiary of Google LLC, is an online community of data scientists and machine learning practitioners. Kaggle allows users to find and publish data sets, explore and build models in a web-based data-science environment, work with other data scientists and machine learning engineers, and enter competitions to solve data science challenges.

Kaggle got its start in 2010 by offering machine learning competitions and now also offers a public data platform, a cloud-based workbench for data science, and Artificial Intelligence education. Its key personnel were Anthony Goldbloom and Jeremy Howard. Nicholas Gruen was founding chair succeeded by Max Levchin. Equity was raised in 2011 valuing the company at $25 million. On 8 March 2017, Google announced that they were acquiring Kaggle.[1][2]

Source: Kaggle

The open dataset, software, and other files accompanying the manuscript "An Open Combinatorial Diffraction Dataset Including Consensus Human and Machine Learning Labels with Quantified Uncertainty for Training New Machine Learning Models," submitted for publication to Integrated Materials and Manufacturing Innovations.Machine learning and autonomy are increasingly prevalent in materials science, but existing models are often trained or tuned using idealized data as absolute ground truths. In actual materials science, "ground truth" is often a matter of interpretation and is more readily determined by consensus. Here we present the data, software, and other files for a study using as-obtained diffraction data as a test case for evaluating the performance of machine learning models in the presence of differing expert opinions. We demonstrate that experts with similar backgrounds can disagree greatly even for something as intuitive as using diffraction to identify the start and end of a phase transformation. We then use a logarithmic likelihood method to evaluate the performance of machine learning models in relation to the consensus expert labels and their variance. We further illustrate this method's efficacy in ranking a number of state-of-the-art phase mapping algorithms. We propose a materials data challenge centered around the problem of evaluating models based on consensus with uncertainty. The data, labels, and code used in this study are all available online at data.gov, and the interested reader is encouraged to replicate and improve the existing models or to propose alternative methods for evaluating algorithmic performance.

Logic synthesis is a challenging and widely-researched combinatorial optimization problem during integrated circuit (IC) design. It transforms a high-level description of hardware in a programming language like Verilog into an optimized digital circuit netlist, a network of interconnected Boolean logic gates, that implements the function. Spurred by the success of ML in solving combinatorial and graph problems in other domains, there is growing interest in the design of ML-guided logic synthesis tools. Yet, there are no standard datasets or prototypical learning tasks defined for this problem domain. Here, we describe OpenABC-D,a large-scale, labeled dataset produced by synthesizing open source designs with a leading open-source logic synthesis tool and illustrate its use in developing, evaluating and benchmarking ML-guided logic synthesis. OpenABC-D has intermediate and final outputs in the form of 870,000 And-Inverter-Graphs (AIGs) produced from 1500 synthesis runs plus labels such as the optimized node counts, and de-lay. We define a generic learning problem on this dataset and benchmark existing solutions for it. The codes related to dataset creation and benchmark models are available athttps://github.com/NYU-MLDA/OpenABC.git.

We introduce PDMX: a Public Domain MusicXML dataset for symbolic music processing, including over 250k musical scores in MusicXML format. PDMX is the largest publicly available, copyright-free MusicXML dataset in existence. PDMX includes genre, tag, description, and popularity metadata for every file.

Description

The Learning Path Index Dataset is a comprehensive collection of byte-sized courses and learning materials tailored for individuals eager to delve into the fields of Data Science, Machine Learning, and Artificial Intelligence (AI), making it an indispensable reference for students, professionals, and educators in the Data Science and AI communities.

This Kaggle Dataset along with the KaggleX Learning Path Index GitHub Repo were created by the mentors and mentees of Cohort 3 KaggleX BIPOC Mentorship Program (between August 2023 and November 2023, also see this). See Credits section at the bottom of the long description.

Inspiration

This dataset was created out of a commitment to facilitate learning and growth within the Data Science, Machine Learning, and AI communities. It started off as an idea at the end of Cohort 2 of the KaggleX BIPOC Mentorship Program brainstorming and feedback session. It was one of the ideas to create byte-sized learning material to help our KaggleX mentees learn things faster. It aspires to simplify the process of finding, evaluating, and selecting the most fitting educational resources.

Context

This dataset was meticulously curated to assist learners in navigating the vast landscape of Data Science, Machine Learning, and AI education. It serves as a compass for those aiming to develop their skills and expertise in these rapidly evolving fields.

The mentors and mentees communicated via Discord, Trello, Google Hangout, etc... to put together these artifacts and made them public for everyone to use and contribute back.

Sources

The dataset compiles data from a curated selection of reputable sources including leading educational platforms such as Google Developer, Google Cloud Skill Boost, IBM, Fast AI, etc. By drawing from these trusted sources, we ensure that the data is both accurate and pertinent. The raw data and other artifacts as a result of this exercise can be found on the GitHub Repo i.e. KaggleX Learning Path Index GitHub Repo.

Content

The dataset encompasses the following attributes:

• Course / Learning Material: The title of the Data Science, Machine Learning, or AI course or learning material.
• Source: The provider or institution offering the course.
• Course Level: The proficiency level, ranging from Beginner to Advanced.
• Type (Free or Paid): Indicates whether the course is available for free or requires payment.
• Module: Specific module or section within the course.
• Duration: The estimated time required to complete the module or course.
• Module / Sub-module Difficulty Level: The complexity level of the module or sub-module.
• Keywords / Tags / Skills / Interests / Categories: Relevant keywords, tags, or categories associated with the course with a focus on Data Science, Machine Learning, and AI.
• Links: Hyperlinks to access the course or learning material directly.

How to contribute to this initiative?

• You can also join us by taking part in the next KaggleX BIPOC Mentorship program (also see this)
• Keep your eyes open on the Kaggle Discussions page and other KaggleX social media channels. Or find us on the Kaggle Discord channel to learn more about the next steps
• Create notebooks from this data
• Create supplementary or complementary data for or from this dataset
• Submit corrections/enhancements or anything else to help improve this dataset so it has a wider use and purpose

License

The Learning Path Index Dataset is openly shared under a permissive license, allowing users to utilize the data for educational, analytical, and research purposes within the Data Science, Machine Learning, and AI domains. Feel free to fork the dataset and make it your own, we would be delighted if you contributed back to the dataset and/or our KaggleX Learning Path Index GitHub Repo as well.

Important Links

• KaggleX BIPOC Mentorship program (also see this)
• KaggleX Learning Path Index Dataset
• KaggleX Learning Path Index GitHub Repo
• New Official Kaggle Discord Server!

Credits

Credits for all the work done to create this Kaggle Dataset and the KaggleX [Learnin...

The purpose of data mining analysis is always to find patterns of the data using certain kind of techiques such as classification or regression. It is not always feasible to apply classification algorithms directly to dataset. Before doing any work on the data, the data has to be pre-processed and this process normally involves feature selection and dimensionality reduction. We tried to use clustering as a way to reduce the dimension of the data and create new features. Based on our project, after using clustering prior to classification, the performance has not improved much. The reason why it has not improved could be the features we selected to perform clustering are not well suited for it. Because of the nature of the data, classification tasks are going to provide more information to work with in terms of improving knowledge and overall performance metrics. From the dimensionality reduction perspective: It is different from Principle Component Analysis which guarantees finding the best linear transformation that reduces the number of dimensions with a minimum loss of information. Using clusters as a technique of reducing the data dimension will lose a lot of information since clustering techniques are based a metric of 'distance'. At high dimensions euclidean distance loses pretty much all meaning. Therefore using clustering as a "Reducing" dimensionality by mapping data points to cluster numbers is not always good since you may lose almost all the information. From the creating new features perspective: Clustering analysis creates labels based on the patterns of the data, it brings uncertainties into the data. By using clustering prior to classification, the decision on the number of clusters will highly affect the performance of the clustering, then affect the performance of classification. If the part of features we use clustering techniques on is very suited for it, it might increase the overall performance on classification. For example, if the features we use k-means on are numerical and the dimension is small, the overall classification performance may be better. We did not lock in the clustering outputs using a random_state in the effort to see if they were stable. Our assumption was that if the results vary highly from run to run which they definitely did, maybe the data just does not cluster well with the methods selected at all. Basically, the ramification we saw was that our results are not much better than random when applying clustering to the data preprocessing. Finally, it is important to ensure a feedback loop is in place to continuously collect the same data in the same format from which the models were created. This feedback loop can be used to measure the model real world effectiveness and also to continue to revise the models from time to time as things change.

Human Body Detection System Using Artificial Intelligence and Machine Learning Deep learning,Open CV, Python, and its Libraries. Basically its a Object Detection system But Enhancing it to Medical Industry.

Curated image privacy data sets

In the article, we trained and evaluated models on the Image Privacy Dataset (IPD) and the PrivacyAlert dataset. The datasets are originally provided by other sources and have been re-organised and curated for this work.

Our curation organises the datasets in a common structure. We updated the annotations and labelled the splits of the data in the annotation file. This avoids having separated folders of images for each data split (training, validation, testing) and allows a flexible handling of new splits, e.g. created with a stratified K-Fold cross-validation procedure. As for the original datasets (PicAlert and PrivacyAlert), we provide the link to the images in bash scripts to download the images. Another bash script re-organises the images in sub-folders with maximum 1000 images in each folder.

Both datasets refer to images publicly available on Flickr. These images have a large variety of content, including sensitive content, seminude people, vehicle plates, documents, private events. Images were annotated with a binary label denoting if the content was deemed to be public or private. As the images are publicly available, their label is mostly public. These datasets have therefore a high imbalance towards the public class. Note that IPD combines two other existing datasets, PicAlert and part of VISPR, to increase the number of private images already limited in PicAlert. Further details in our corresponding https://doi.org/10.48550/arXiv.2503.12464" target="_blank" rel="noopener">publication.

List of datasets and their original source:

• PicAlert [Images occupy 2.4 GB]
• VISPR [Images occupy 49.7 GB]
• PrivacyAlert [Images occupy 1 GB]

Notes:

• For PicAlert and PrivacyAlert, only urls to the original locations in Flickr are available in the Zenodo record
• Collector and authors of the PrivacyAlert dataset selected the images from Flickr under Public Domain license
• Owners of the photos on Flick could have removed the photos from the social media platform
• Running the bash scripts to download the images can incur in the "429 Too Many Requests" status code

Pre-computed visual entitities

Some of the models run their pipeline end-to-end with the images as input, whereas other models require different or additional inputs. These inputs include the pre-computed visual entities (scene types and object types) represented in a graph format, e.g. for a Graph Neural Network. Re-using these pre-computed visual entities allows other researcher to build new models based on these features while avoiding re-computing the same on their own or for each epoch during the training of a model (faster training).

For each image of each dataset, namely PrivacyAlert, PicAlert, and VISPR, we provide the predicted scene probabilities as a .csv file , the detected objects as a .json file in COCO data format, and the node features (visual entities already organised in graph format with their features) as a .json file. For consistency, all the files are already organised in batches following the structure of the images in the datasets folder. For each dataset, we also provide the pre-computed adjacency matrix for the graph data.

Note: IPD is based on PicAlert and VISPR and therefore IPD refers to the scene probabilities and object detections of the other two datasets. Both PicAlert and VISPR must be downloaded and prepared to use IPD for training and testing.

Further details on downloading and organising data can be found in our GitHub repository: https://github.com/graphnex/privacy-from-visual-entities (see ARTIFACT-EVALUATION.md#pre-computed-visual-entitities-)

Enquiries, questions and comments

If you have any enquiries, question, or comments, or you would like to file a bug report or a feature request, use the issue tracker of our GitHub repository.

General

For more details and the most up-to-date information please consult our project page: https://kainmueller-lab.github.io/fisbe.

Summary

A new dataset for neuron instance segmentation in 3d multicolor light microscopy data of fruit fly brains

30 completely labeled (segmented) images

71 partly labeled images

altogether comprising ∼600 expert-labeled neuron instances (labeling a single neuron takes between 30-60 min on average, yet a difficult one can take up to 4 hours)

To the best of our knowledge, the first real-world benchmark dataset for instance segmentation of long thin filamentous objects

A set of metrics and a novel ranking score for respective meaningful method benchmarking

An evaluation of three baseline methods in terms of the above metrics and score

Abstract

Instance segmentation of neurons in volumetric light microscopy images of nervous systems enables groundbreaking research in neuroscience by facilitating joint functional and morphological analyses of neural circuits at cellular resolution. Yet said multi-neuron light microscopy data exhibits extremely challenging properties for the task of instance segmentation: Individual neurons have long-ranging, thin filamentous and widely branching morphologies, multiple neurons are tightly inter-weaved, and partial volume effects, uneven illumination and noise inherent to light microscopy severely impede local disentangling as well as long-range tracing of individual neurons. These properties reflect a current key challenge in machine learning research, namely to effectively capture long-range dependencies in the data. While respective methodological research is buzzing, to date methods are typically benchmarked on synthetic datasets. To address this gap, we release the FlyLight Instance Segmentation Benchmark (FISBe) dataset, the first publicly available multi-neuron light microscopy dataset with pixel-wise annotations. In addition, we define a set of instance segmentation metrics for benchmarking that we designed to be meaningful with regard to downstream analyses. Lastly, we provide three baselines to kick off a competition that we envision to both advance the field of machine learning regarding methodology for capturing long-range data dependencies, and facilitate scientific discovery in basic neuroscience.

Dataset documentation:

We provide a detailed documentation of our dataset, following the Datasheet for Datasets questionnaire:

FISBe Datasheet

Our dataset originates from the FlyLight project, where the authors released a large image collection of nervous systems of ~74,000 flies, available for download under CC BY 4.0 license.

Files

fisbe_v1.0_{completely,partly}.zip

contains the image and ground truth segmentation data; there is one zarr file per sample, see below for more information on how to access zarr files.

fisbe_v1.0_mips.zip

maximum intensity projections of all samples, for convenience.

sample_list_per_split.txt

a simple list of all samples and the subset they are in, for convenience.

view_data.py

a simple python script to visualize samples, see below for more information on how to use it.

dim_neurons_val_and_test_sets.json

a list of instance ids per sample that are considered to be of low intensity/dim; can be used for extended evaluation.

Readme.md

general information

How to work with the image files

Each sample consists of a single 3d MCFO image of neurons of the fruit fly.For each image, we provide a pixel-wise instance segmentation for all separable neurons.Each sample is stored as a separate zarr file (zarr is a file storage format for chunked, compressed, N-dimensional arrays based on an open-source specification.").The image data ("raw") and the segmentation ("gt_instances") are stored as two arrays within a single zarr file.The segmentation mask for each neuron is stored in a separate channel.The order of dimensions is CZYX.

We recommend to work in a virtual environment, e.g., by using conda:

conda create -y -n flylight-env -c conda-forge python=3.9conda activate flylight-env

How to open zarr files

Install the python zarr package:

pip install zarr

Opened a zarr file with:

import zarrraw = zarr.open(, mode='r', path="volumes/raw")seg = zarr.open(, mode='r', path="volumes/gt_instances")

optional:import numpy as npraw_np = np.array(raw)

Zarr arrays are read lazily on-demand.Many functions that expect numpy arrays also work with zarr arrays.Optionally, the arrays can also explicitly be converted to numpy arrays.

How to view zarr image files

We recommend to use napari to view the image data.

Install napari:

pip install "napari[all]"

Save the following Python script:

import zarr, sys, napari

raw = zarr.load(sys.argv[1], mode='r', path="volumes/raw")gts = zarr.load(sys.argv[1], mode='r', path="volumes/gt_instances")

viewer = napari.Viewer(ndisplay=3)for idx, gt in enumerate(gts): viewer.add_labels( gt, rendering='translucent', blending='additive', name=f'gt_{idx}')viewer.add_image(raw[0], colormap="red", name='raw_r', blending='additive')viewer.add_image(raw[1], colormap="green", name='raw_g', blending='additive')viewer.add_image(raw[2], colormap="blue", name='raw_b', blending='additive')napari.run()

Execute:

python view_data.py /R9F03-20181030_62_B5.zarr

Metrics

S: Average of avF1 and C

avF1: Average F1 Score

C: Average ground truth coverage

clDice_TP: Average true positives clDice

FS: Number of false splits

FM: Number of false merges

tp: Relative number of true positives

For more information on our selected metrics and formal definitions please see our paper.

Baseline

To showcase the FISBe dataset together with our selection of metrics, we provide evaluation results for three baseline methods, namely PatchPerPix (ppp), Flood Filling Networks (FFN) and a non-learnt application-specific color clustering from Duan et al..For detailed information on the methods and the quantitative results please see our paper.

License

The FlyLight Instance Segmentation Benchmark (FISBe) dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

Citation

If you use FISBe in your research, please use the following BibTeX entry:

@misc{mais2024fisbe, title = {FISBe: A real-world benchmark dataset for instance segmentation of long-range thin filamentous structures}, author = {Lisa Mais and Peter Hirsch and Claire Managan and Ramya Kandarpa and Josef Lorenz Rumberger and Annika Reinke and Lena Maier-Hein and Gudrun Ihrke and Dagmar Kainmueller}, year = 2024, eprint = {2404.00130}, archivePrefix ={arXiv}, primaryClass = {cs.CV} }

Acknowledgments

We thank Aljoscha Nern for providing unpublished MCFO images as well as Geoffrey W. Meissner and the entire FlyLight Project Team for valuablediscussions.P.H., L.M. and D.K. were supported by the HHMI Janelia Visiting Scientist Program.This work was co-funded by Helmholtz Imaging.

Changelog

There have been no changes to the dataset so far.All future change will be listed on the changelog page.

Contributing

If you would like to contribute, have encountered any issues or have any suggestions, please open an issue for the FISBe dataset in the accompanying github repository.

All contributions are welcome!