Iris Flower Dataset

This is a classic and very widely used dataset in machine learning and statistics, often serving as a first dataset for classification problems. Introduced by the British statistician and biologist Ronald Fisher in his 1936 paper "The use of multiple measurements in taxonomic problems," it is a foundational resource for learning classification algorithms.

Overview:

The dataset contains measurements for 150 samples of iris flowers. Each sample belongs to one of three species of iris:

• Iris setosa
• Iris versicolor
• Iris virginica

For each flower, four features were measured:

• Sepal length (in cm)
• Sepal width (in cm)
• Petal length (in cm)
• Petal width (in cm)

The goal is typically to build a model that can classify iris flowers into their correct species based on these four features.

File Structure:

The dataset is usually provided as a single CSV (Comma Separated Values) file, often named iris.csv or similar. This file typically contains the following columns:

1. sepal_length (cm): Numerical. The length of the sepal of the iris flower.
2. sepal_width (cm): Numerical. The width of the sepal of the iris flower.
3. petal_length (cm): Numerical. The length of the petal of the iris flower.
4. petal_width (cm): Numerical. The width of the petal of the iris flower.
5. species: Categorical. The species of the iris flower (either 'setosa', 'versicolor', or 'virginica'). This is the target variable for classification.

Content of the Data:

The dataset contains an equal number of samples (50) for each of the three iris species. The measurements of the sepal and petal dimensions vary between the species, allowing for their differentiation using machine learning models.

How to Use This Dataset:

1. Download the iris.csv file.
2. Load the data using libraries like Pandas in Python.
3. Explore the data through visualization and statistical analysis to understand the relationships between the features and the different species.
4. Build classification models (e.g., Logistic Regression, Support Vector Machines, Decision Trees, K-Nearest Neighbors) using the sepal and petal measurements as features and the 'species' column as the target variable.
5. Evaluate the performance of your model using appropriate metrics (e.g., accuracy, precision, recall, F1-score).
6. The dataset is small and well-behaved, making it excellent for learning and experimenting with various classification techniques.

Citation:

When using the Iris dataset, it is common to cite Ronald Fisher's original work:

Fisher, R. A. (1936). The use of multiple measurements in taxonomic problems. Annals of Eugenics, 7(2), 179-188.

Data Contribution:

Thank you for providing this classic and fundamental dataset to the Kaggle community. The Iris dataset remains an invaluable resource for both beginners learning the basics of classification and experienced practitioners testing new algorithms. Its simplicity and clear class separation make it an ideal starting point for many data science projects.

If you find this dataset description helpful and the dataset itself useful for your learning or projects, please consider giving it an upvote after downloading. Your appreciation is valuable!

Iris

The following code can be used to load the dataset from its stored location at NERSC. You may also access this code via a NERSC-hosted Jupyter notebook here.

Iris data loader

import pandas as pd

iris_dat = pd.read_csv('/global/cfs/cdirs/dasrepo/www/ai_ready_datasets/iris/data/iris.csv')

If you would like to download the data, visit the following link: https://portal.nersc.gov/cfs/dasrepo/ai_ready_datasets/iris/data/

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.

28 Global import shipment records of Iris Retractor with prices, volume & current Buyer's suppliers relationships based on actual Global export trade database.

Eximpedia Export import trade data lets you search trade data and active Exporters, Importers, Buyers, Suppliers, manufacturers exporters from over 209 countries

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.

Paradoxical tuberculosis-associated immune reconstitution inflammatory syndrome (TB-IRIS) is an aberrant inflammatory response occurring in a subset of TB-HIV co-infected patients initiating anti-retroviral therapy (ART). Here, we examined monocyte activation by prospectively quantitating pro-inflammatory plasma markers and monocyte subsets in TB-HIV co-infected patients from a South Indian cohort at baseline and following ART initiation at the time of IRIS, or at equivalent time points in non-IRIS controls. Pro-inflammatory biomarkers of innate and myeloid cell activation were increased in plasma of IRIS patients pre-ART and at the time of IRIS; this association was confirmed in a second cohort in South Africa. Increased expression of these markers correlated with elevated antigen load as measured by higher sputum culture grade and shorter duration of anti-TB therapy. Phenotypic analysis revealed the frequency of CD14++CD16− monocytes was an independent predictor of TB-IRIS, and was closely associated with plasma levels of CRP, TNF, IL-6 and tissue factor during IRIS. In addition, production of inflammatory cytokines by monocytes was higher in IRIS patients compared to controls pre-ART. These data point to a major role of mycobacterial antigen load and myeloid cell hyperactivation in the pathogenesis of TB-IRIS, and implicate monocytes and monocyte-derived cytokines as potential targets for TB-IRIS prevention or treatment.

Llp Iris Dreams Company Export Import Records. Follow the Eximpedia platform for HS code, importer-exporter records, and customs shipment details.

Iris Tech North America Llc Company Export Import Records. Follow the Eximpedia platform for HS code, importer-exporter records, and customs shipment details.

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.

Eximpedia Export import trade data lets you search trade data and active Exporters, Importers, Buyers, Suppliers, manufacturers exporters from over 209 countries

Iris Global Dmcc Company Export Import Records. Follow the Eximpedia platform for HS code, importer-exporter records, and customs shipment details.

Eximpedia Export import trade data lets you search trade data and active Exporters, Importers, Buyers, Suppliers, manufacturers exporters from over 209 countries

89 Global import shipment records of Iris Retractor And HSN Code 9018 with prices, volume & current Buyer's suppliers relationships based on actual Global export trade database.

752 Global exporters importers export import shipment records of Flexible iris retractor with prices, volume & current Buyer's suppliers relationships based on actual Global export trade database.

Details related to access to the data

• Contact person

Please contact Iris Groen (i.i.a.groen@uva.nl, https://orcid.org/0000-0002-5536-6128) for more information.

Please see the following papers for more details on the data collection and preprocessing:

Groen IIA, Piantoni G, Montenegro S, Flinker A, Devore S, Devinsky O, Doyle W, Dugan P, Friedman D, Ramsey N, Petridou N, Winawer JA (2022) Temporal dynamics of neural responses in human visual cortex. The Journal of Neuroscience 42(40):7562-7580 (https://doi.org/10.1523/JNEUROSCI.1812-21.2022)

Yuasa K, Groen IIA, Piantoni G, Montenegro S, Flinker A, Devore S, Devinsky O, Doyle W, Dugan P, Friedman D, Ramsey N, Petridou N, Winawer JA. Precise Spatial Tuning of Visually Driven Alpha Oscillations in Human Visual Cortex. eLife12:RP90387 https://doi.org/10.7554/eLife.90387.1

Brands AM, Devore S, Devinsky O, Doyle W, Flinker A, Friedman D, Dugan P, Winawer JA, Groen IIA (2024). Temporal dynamics of short-term neural adaptation in human visual cortex. https://doi.org/10.1101/2023.09.13.557378

• Practical information to access the data

Processed data and model fits reported in Groen et al., (2022) are available in derivatives/Groenetal2022TemporalDynamicsECoG as matlab .mat files. Matlab code to load, process and plot these data (including 3D renderings of the participant's surface reconstructions and electrode positions) is available in https://github.com/WinawerLab/ECoG_utils and https://github.com/irisgroen/temporalECoG. These repositories have dependencies on other Matlab toolboxes (e.g., FieldTrip). See instructions on Github for relevant links and guidelines.

Processed data and model fits reported in Yuasa et al., (2023) are available in the Github repositories described in the paper.

Processed data and model fits reported in Brands et al., (2024) are available in derivatives/Brandsetal2024TemporalAdaptationECoGCategories as python .py files. Python code to process and analyze these data is available in the Github repositories described in the paper.

Overview

• Project name

Visual ECoG dataset

• Years that the project ran

Data were collected between 2017-2020. Exact recording dates have been scrubbed for anonymization purposes.

• Brief overview of the tasks in the experiment

Participants sub-p01 to sub-p11 viewed grayscale visual pattern stimuli that were varied in temporal or spatial properties. Participans sub-p11 to sub-p14 additionally saw color images of different image classes (faces, bodies, buildings, objects, scenes, and scrambled) that were varied in temporal properties. See 'Independent Variables' below for more details.

In all tasks, participants were instructed to fixate a cross or point in the center of the screen and monitor it for a color change, i.e. to perform a stimulus-orthogonal task (see the task-specific _events.json files, e.g., task-prf_events.json, for further details).

• Description of the contents of the dataset

The data consists of cortical iEEG recordings in 14 epilepsy patients in response to visual stimulation. Patients were implanted with standard clinical surface (grid) and depth electrodes. Two patients were additionally implanted with a high-density research grid. In addition to the ieeg recordings, pre-implantation MRI T1 scans are provided for the purpose of localizing electrodes. Participants performed a varying number of tasks and runs.

• Independent variables

The data are divided in 6 different sets of stimulus types or events:

1. prf: grayscale, oriented bar stimuli consisting of curved, band-pass filtered lines that were swept across the screen (up to (~16 degree of visual angle) in a fixed order for the purpose of estimating spatial population receptive fields (pRFs).
2. spatialpattern: grayscale, centrally presented pattern stimuli (~16 degree of visual angle diameter) consisting of curved, band-pass filtered lines that were systematically varied in level of contrast and density, as well as various oriented grating stimuli.
3. temporalpattern: grayscale, centrally presented pattern stimuli (~16 degree of visual angle diameter) consisting of curved, band-pass filtered lines that were systematically varied in temporal duration and interval.
4. soc: combination of the spatialpattern and temporalpattern stimuli.
5. sixcatloctemporal: color images of six stimulus classes: faces, bodies (hands/feet only), buildings, objects, scenes and scrambled, systematically varied in temporal duration and interval, whereby interval stimuli consisted of direct repeats of the identical image.
6. sixcatlocisidiff/sixcatlocdiffisi: color images of six stimulus classes: faces, bodies (hands/feet only), buildings, objects, scenes and scrambled, systematically varied in temporal duration and interval, whereby the first interval stimulus was followed by images from either the same or a different category (but not the identical image).

Participant-, task- and run-specific stimuli are provided in the /stimuli folder as matlab .mat files.

• Dependent variables

The main BIDS folder contains the raw voltage data, split up in individual task runs. The /derivatives/ECoGCAR folder contains common-average-referenced version of the data. The /derivatives/ECoGBroadband folder contains time-varying broadband responses estimated by band-pass filtering the common-average-referenced voltage data and taking the average power envelope. The /derivatives/ECoGPreprocessed folder contains epoched trials used in Brands et al., (2024). The /derivatives/freesurfer folder contains surface reconstructions of each participant's T1, along with retinotopic atlas files. The /derivatives/Groen2022TemporalDynamicsECoG contains preprocessed data and model fits that can be used to reproduce the results reported in Groen et al., (2022). The /derivatives/Brands2024TemporalAdaptationECoG contains preprocessed data and model fits that can be used to reproduce the results reported in Brands et al., (2024).

• Quality assessment of the data

Data quality and number of trials per subjects varies considerably across patients, for various reasons.

First, for each recording session, attempts were made to optimize the environment for running visual experiments; e.g. room illumination was stabilized as much as possible by closing blinds when available, the visual display was calibrated (for most patients), and interference from medical staff or visitors was minimized. However, it was not possible to equate this with great precision across patients and sessions/runs.

Second, implantations were determined based on clinical needs and electrode locations therefore vary across participants. The strength and robustness of the neural responses varies greatly with the electrode location (e.g. early vs higher-level visual cortex), as well as with uncontrolled factors such as how well the electrode made contact with the cortex and whether it was primarily situated on grey matter (surface/grid electrodes) or could be located in white matter (some depth electrodes). Electrodes that were marked as containing epileptic activity by clinicians, or that did not have good signal based on visual inspection of the raw data, are marked as 'bad' in the channels.tsv files.

Third, patients varied greatly in their cognitive abilities and mental/medical state, which affected their ability to follow task instructions, e.g. to remain alert and fixation. Some patients were able to perform repeated runs of multiple tasks across multiple sessions, while others only managed to do a few runs.

All patients included in this dataset have sufficiently good responses in some electrodes/tasks as judged by Groen et al., (2022) and Brands et al., (2024). However, when using this dataset to address further research questions, it is advisable to set stringent requirements on electrode and trial selection. See Groen et al., (2022) and associated code repository for an example preprocessing pipeline that selected for robust visual responses to temporally- and contrast-varying stimuli.

Methods

• Subjects

All participants were intractable epilepsy patients who were undergoing ECoG for the purpose of monitoring seizures. Participants were included if their implantation covered parts of visual cortex and if they consented to participate in research.

• Apparatus

Data were collected in a clinical setting, i.e. at bedside in the patient's hospital room. Information about iEEG recording apparatus is provided the meta data for each patient. Information about the visual stimulation equipment and behavioral response recordings are provided in Groen et al., (2022), Yuasa et al., (2023) and Brands et al., (2024).

• Experimental location

Data were collected at NYU University Langone Hospital (New York, USA) or at University Medical Center Utrecht (The Netherlands).

• Missing data

Stimulus files are missing for a few runs of sub-02. These are marked as N/A in the associated event files.

Notes

Further participant-specific notes:

• For sub-03 and sub-04 the spatial pattern and temporal pattern stimuli are combined in the soc task runs, for the remaining participants these are split across the spatialpattern and temporalpattern task runs.

• The pRF task from sub-04 has different prf parameters (bar duration and gap).

• The first two runs of the pRF task from sub-05 are not of good quality (participant repeatedly broke fixation). In addition, the triggers in all pRF runs from sub-05 are not correct due to a stimulus coding problem and will need to be re-interpolated if one wishes to use these data.

• Participants sub-10 and sub-11 have high density grids in addition to clinical grids.

• Note that all stimuli and stimulus parameters can be found in the participant-specific stimulus *.mat files.

Iris Dyes And Pigments Company Export Import Records. Follow the Eximpedia platform for HS code, importer-exporter records, and customs shipment details.

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.

Successful Short-Term Volcanic Eruption Forecasting Using Seismic Features, Suplementary Material by Rey-Devesa (1,2), Benítez (3), Prudencio, Ligdamis Gutiérrez (1,2), Cortés (1,2), Titos (3), Koulakov (4,5), Zuccarello (6) and Ibáñez (1,2).
Institutions associated: (1) Department of Theoretical Physics and Cosmos. Science Faculty. Avd. Fuentenueva s/n. University of Granada. 18071. Granada. Spain. (2) Andalusian Institute of Geophysiscs. Campus de Cartuja. University of Granada. C/Profesor Clavera 12. 18071. Granada. Spain. (3) Department of Signal Theory, Telematics and Communication. University of Granada. Informatics and Telecommunication School. 18071. Granada. Spain. (4) Trofimuk Institute of Petroleum Geology and Geophysics SB RAS, Prospekt Koptyuga, 3, 630090 Novosibirsk, Russia (5) Institute of the Earth’s Crust SB RAS, Lermontova 128, Irkutsk, Russia (6) Istituto Nazionale di Geofisica e Vulcanologia, Sezione di Pisa (INGV-Pisa), via Cesare Battisti, 53, 56125, Pisa, Italy.
Acknowledgment: This study was partially supported by the Spanish FEMALE project (PID2019-106260GB-I00).
P. Rey-Devesa was funded by the Ministerio de Ciencia e Innovación del Gobierno de España (MCIN),
Agencia Estatal de Investigación (AEI), Fondo Social Europeo (FSE),
and Programa Estatal de Promoción del Talento y su Empleabilidad en I+D+I Ayudas para contratos predoctorales para la formación de doctores 2020 (PRE2020-092719).
Ivan Koulakov was supported by the Russian Science Foundation (Grant No. 20-17-00075).
Luciano Zuccarello was supported by the INGV Pianeta Dinamico 2021 Tema 8 SOME project (grant no. CUP D53J1900017001)
funded by the Italian Ministry of University and Research
“Fondo finalizzato al rilancio degli investimenti delle amministrazioni centrali dello Stato e allo sviluppo del Paese, legge 145/2018”.
English language editing was performed by Tornillo Scientific, UK.
Data availability statement: 1.- Seismic data from Kilauea, Augustine, Bezymianny (2007), and Mount St. Helens are available from the IRIS data repository (http://ds.iris.edu/seismon/index.phtml).
(An example of the Python code to access the data is described below.)
2.- Seismic data from Bezymianny (2017-2018) are available from Ivan Koulakov (ivan.science@gmail.com) upon request.
3.- Seismic data from Mt. Etna are available from INGV-Italy upon request (http://terremoti.ingv.it/en/help),
also available from the Zenodo data repository (https://doi.org/10.5281/zenodo.6849621). Access code in Python to download the records of Kilauea, Augustine and Mount St. Helens volcanoes, from the IRIS data repository. '''To access the raw signals please first install ObsPy and then execute following commands in a python console: ''' Example: from obspy.core import UTCDateTime
from obspy.clients.fdsn import Client
import obspy.io.mseed
client = Client('IRIS')
t1 = UTCDateTime('2006-01-10T00:00:00')
t2 = UTCDateTime('2006-01-12T00:00:00')
raw_data = client.get_waveforms(
network='AV',
station='AUH',
location='',
channel='HHZ',
starttime=t1,
endtime=t2) '''To further download station information execute: ''' xml = client.get_stations(network='AV',station='AUH',
channel='HHZ',starttime=t1,endtime=t2,level='response') ''' 'To scale the data using the station’s meta-data: ''' data = raw_data.remove_response(inventory=xml) ''' To filter, trim and plot the data execute: ''' data.write("Augustine.mseed", format="MSEED") data.filter('bandpass',freqmin=1.0,freqmax=20)
data.trim(t1+60,t2-60)
data.plot() Contents: 6 different Matlab codes. The principal code is called FeatureExtraction.
The codes rsac.m and ReadMSEEDFast.m are for reading different format of data. (Not developed by the group)
Seismic Data from Mt. Etna for using as an example.

BackgroundAfter initiation of combination antiretroviral treatment (cART), HIV-1/tuberculosis coinfected patients are at high risk of developing tuberculosis-associated immune reconstitution inflammatory syndrome (TB-IRIS). MicroRNAs, small molecules of approximately 22 nucleotides, which regulate post-transcriptional gene expression and their profile has been proposed as a biomarker for many diseases. We tested whether the microRNA profile could be a predictive biomarker for TB-IRIS.MethodsTwenty-six selected microRNAs involved in the regulation of the innate immune response were investigated. Free plasmatic and microRNA-derived exosomes were measured by flow cytometry. The plasma from 74 HIV-1+TB+ individuals (35 IRIS and 39 non-IRIS) at the time of the diagnosis and before any treatment (baseline) of CAMELIA trial (ANRS1295-CIPRA KH001-DAIDS-ES ID10425); 15 HIV+TB− and 23 HIV−TB+, both naïve of any treatment; and 20 HIV−TB− individuals as controls were analysed.ResultsAt baseline, both IRIS and non-IRIS HIV+/TB+ individuals had similar demographic and clinical characteristics, including sex, age, body mass index, very low CD4+ cell counts (27 cells/mm3), and plasma HIV RNA load levels (5.76 log copies/ml). Twenty out of 26 plasmatic-microRNAs tested were no different between IRIS and controls. Twelve of the 26 tested microRNAs showed statistically significant differences between IRIS and non-IRIS patients (p-values ranging from p