radMLBench 1.0__py3-none-any.whl

radMLBench/__init__.py

@@ -0,0 +1,32 @@
+# -*- coding: utf-8 -*-
+
+"""
+RadMLBench was developed at the University Hospital in Essen, Germany.
+For questions use github issues or write an email (aydin.demircioglu@uk-essen.de).
+
+RadMLBench is partially based on the PMLB (Penn Machine Learning Benchmarks), see https://epistasislab.github.io/pmlb/
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of this software
+and associated documentation files (the "Software"), to deal in the Software without restriction,
+including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense,
+and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so,
+subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all copies or substantial
+portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT
+LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+"""
+
+from .radMLBench import (
+    listDatasets,
+    loadData,
+    getCVSplits,
+    getMetaData)
+
+
+#

radMLBench/metadata.yaml

@@ -0,0 +1,976 @@
+Ahn2021:
+  BrokenFeatures: 0
+  ClassBalance: 55
+  Dimensionality: 0.96
+  Missings: 0
+  authors: Choi et al.
+  dataset: Ahn2021
+  download_source: https://data.mendeley.com/datasets/9mpy8mc6hn/1
+  modality: MRI
+  nFeatures: 108
+  nInstances: 114
+  outcome: MGMT mutation status
+  pathology: Glioblastoma
+  publication_doi: https://doi.org/10.1016/j.ejrad.2019.108642
+  publication_title: 'Analysis of heterogeneity of peritumoral T2 hyperintensity in
+    patients with pretreatment glioblastoma: Prognostic value of MRI-based radiomics'
+  remarks: null
+  software: PyRadiomics
+  year: 2021
+Arita2018:
+  BrokenFeatures: 0
+  ClassBalance: 66
+  Dimensionality: 4.08
+  Missings: 168
+  authors: Arita et al.
+  dataset: Arita2018
+  download_source: https://www.nature.com/articles/s41598-018-30273-4#Sec18
+  modality: MRI
+  nFeatures: 683
+  nInstances: 168
+  outcome: IDH mutation status
+  pathology: Glioma
+  publication_doi: https://doi.org/10.1038/s41598-018-30273-4
+  publication_title: Lesion location implemented magnetic resonance imaging radiomics
+    for predicting IDH and TERT promoter mutations in grade II/III gliomas
+  remarks: null
+  software: Inhouse
+  year: 2018
+BraTS-2021:
+  BrokenFeatures: 0
+  ClassBalance: 52
+  Dimensionality: 7.04
+  Missings: 0
+  authors: Flanders et al.
+  dataset: BraTS-2021
+  download_source: https://radiomics.uk/
+  modality: MRI
+  nFeatures: 4060
+  nInstances: 577
+  outcome: MGMT mutation status
+  pathology: Glioblastoma
+  publication_doi: https://doi.org/10.48550/arXiv.2107.02314
+  publication_title: The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation
+    and Radiogenomic Classification
+  remarks: Original dataset at https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification
+  software: PyRadiomics
+  year: 2021
+Brancato2023:
+  BrokenFeatures: 0
+  ClassBalance: 60
+  Dimensionality: 41.07
+  Missings: 0
+  authors: Castaldo et al.
+  dataset: Brancato2023
+  download_source: https://data.mendeley.com/datasets/dvysdynchh/1
+  modality: MRI
+  nFeatures: 2380
+  nInstances: 58
+  outcome: Clinical significance
+  pathology: Prostate cancer
+  publication_doi: https://doi.org/10.3390/jcm12010140
+  publication_title: A Framework of Analysis to Facilitate the Harmonization of Multicenter
+    Radiomic Features in Prostate Cancer
+  remarks: Only center 2 was used, since no labels were available for center 1
+  software: PyRadiomics
+  year: 2023
+C4KC-KiTS:
+  BrokenFeatures: 0
+  ClassBalance: 66
+  Dimensionality: 4.53
+  Missings: 0
+  authors: Heller et al.
+  dataset: C4KC-KiTS
+  download_source: https://radiomics.uk/
+  modality: CT
+  nFeatures: 315
+  nInstances: 70
+  outcome: Histological subtype
+  pathology: Kidney tumor
+  publication_doi: https://doi.org/10.1016/j.media.2020.101821
+  publication_title: 'The state of the art in kidney and kidney tumor segmentation
+    in contrast-enhanced CT imaging: Results of the KiTS19 challenge'
+  remarks: Original dataset at https://www.cancerimagingarchive.net/collection/c4kc-kits.
+  software: PyRadiomics
+  year: 2021
+Colorectal-Liver-Metastases:
+  BrokenFeatures: 0
+  ClassBalance: 84
+  Dimensionality: 5.86
+  Missings: 0
+  authors: Simpson et al.
+  dataset: Colorectal-Liver-Metastases
+  download_source: https://radiomics.uk/
+  modality: CT
+  nFeatures: 525
+  nInstances: 90
+  outcome: Survival (at 10 years)
+  pathology: Colorectal cancer
+  publication_doi: https://doi.org/10.1038/s41597-024-02981-2
+  publication_title: Preoperative CT and survival data for patients undergoing resection
+    of colorectal liver metastases
+  remarks: Original dataset at https://www.cancerimagingarchive.net/collection/colorectal-liver-metastases.
+    Only patient that died were included.
+  software: PyRadiomics
+  year: 2024
+Dai2023:
+  BrokenFeatures: 0
+  ClassBalance: 26
+  Dimensionality: 7.17
+  Missings: 0
+  authors: Dai et al.
+  dataset: Dai2023
+  download_source: https://peerj.com/articles/16230/#supplemental-information
+  modality: CT
+  nFeatures: 851
+  nInstances: 119
+  outcome: Thrombocytopenia presence
+  pathology: Gastrointestinal malignancies
+  publication_doi: https://doi.org/10.7717/peerj.16230
+  publication_title: 'A contrast-enhanced CT-based whole-spleen radiomics signature
+    for early prediction of oxaliplatin-related thrombocytopenia in patients with
+    gastrointestinal malignancies: a retrospective study'
+  remarks: null
+  software: PyRadiomics
+  year: 2023
+Deng2023:
+  BrokenFeatures: 1
+  ClassBalance: 36
+  Dimensionality: 0.87
+  Missings: 0
+  authors: Deng et al.
+  dataset: Deng2023
+  download_source: https://github.com/BboyT/BM_NSCLC_subpathology
+  modality: MRI
+  nFeatures: 225
+  nInstances: 261
+  outcome: Pathological diagnosis
+  pathology: NSCLC
+  publication_doi: https://doi.org/10.1007/s13246-023-01300-0
+  publication_title: 'MRI radiomics for brain metastasis sub-pathology classification
+    from non-small cell lung cancer: a machine learning, multicenter study'
+  remarks: All data were merged. Attached python applies CV 'wrongly' on internal
+    data, but there is external data, so test results are unbiased.
+  software: PyRadiomics
+  year: 2023
+Dong2022:
+  BrokenFeatures: 0
+  ClassBalance: 49
+  Dimensionality: 3.06
+  Missings: 0
+  authors: Dong et al.
+  dataset: Dong2022
+  download_source: https://peerj.com/articles/14127/#supplemental-information
+  modality: CT
+  nFeatures: 851
+  nInstances: 279
+  outcome: Tuberculosis granuloma presence
+  pathology: Lung cancer
+  publication_doi: https://doi.org/10.7717/peerj.14127
+  publication_title: Radiomics combined with clinical features in distinguishing non-calcifying
+    tuberculosis granuloma and lung adenocarcinoma in small pulmonary nodules
+  remarks: null
+  software: PyRadiomics (FAE)
+  year: 2022
+Fusco2022:
+  BrokenFeatures: 0
+  ClassBalance: 61
+  Dimensionality: 3.59
+  Missings: 0
+  authors: Fusco et al.
+  dataset: Fusco2022
+  download_source: https://zenodo.org/records/6344730
+  modality: MRI
+  nFeatures: 192
+  nInstances: 54
+  outcome: Malignancy presence
+  pathology: Breast cancer
+  publication_doi: https://doi.org/10.3390/curroncol29030159
+  publication_title: Radiomic and Artificial Intelligence Analysis with Textural Metrics
+    Extracted by Contrast-Enhanced Mammography and Dynamic Contrast Magnetic Resonance
+    Imaging to Detect Breast Malignant Lesions
+  remarks: null
+  software: TextureToolbox
+  year: 2022
+Granata2021:
+  BrokenFeatures: 3
+  ClassBalance: 47
+  Dimensionality: 6.61
+  Missings: 0
+  authors: Granata et al.
+  dataset: Granata2021
+  download_source: https://zenodo.org/records/5162861
+  modality: CT
+  nFeatures: 580
+  nInstances: 88
+  outcome: Overall survival (at 32 months)
+  pathology: Lung adenocarcinoma
+  publication_doi: https://doi.org/10.3390/cancers13163992
+  publication_title: Preliminary Report on Computed Tomography Radiomics Features
+    as Biomarkers to Immunotherapy Selection in Lung Adenocarcinoma Patients
+  remarks: No dead status/event is given.
+  software: HealthMyne (Commercial)
+  year: 2021
+Granata2024:
+  BrokenFeatures: 0
+  ClassBalance: 75
+  Dimensionality: 16.73
+  Missings: 0
+  authors: Granata et al.
+  dataset: Granata2024
+  download_source: https://zenodo.org/records/10464602
+  modality: MRI
+  nFeatures: 851
+  nInstances: 51
+  outcome: Tumor budding presence
+  pathology: Colorectal cancer
+  publication_doi: https://doi.org/10.3390/diagnostics14020152
+  publication_title: Machine Learning and Radiomics Analysis for Tumor Budding Prediction
+    in Colorectal Liver Metastases Magnetic Resonance Imaging Assessment
+  remarks: null
+  software: PyRadiomics
+  year: 2024
+HCC-TACE-Seg:
+  BrokenFeatures: 0
+  ClassBalance: 14
+  Dimensionality: 5.02
+  Missings: 0
+  authors: Moawad et al.
+  dataset: HCC-TACE
+  download_source: https://radiomics.uk/
+  modality: CT
+  nFeatures: 420
+  nInstances: 84
+  outcome: Lymph node metastasis presence
+  pathology: Hepatocellular cancer
+  publication_doi: https://doi.org/10.1038/s41597-023-01928-3
+  publication_title: Multimodality annotated hepatocellular carcinoma data set including
+    pre- and post-TACE with imaging segmentation
+  remarks: Original data at https://www.cancerimagingarchive.net/collection/hcc-tace-seg/.
+  software: PyRadiomics
+  year: 2023
+HNSCC:
+  BrokenFeatures: 0
+  ClassBalance: 27
+  Dimensionality: 1.15
+  Missings: 0
+  authors: Grossberg et al.
+  dataset: HNSCC
+  download_source: https://radiomics.uk/
+  modality: CT
+  nFeatures: 105
+  nInstances: 93
+  outcome: Lymph node metastasis presence
+  pathology: Head-and-neck cancer
+  publication_doi: https://doi.org/10.1038/sdata.2018.173
+  publication_title: Imaging and Clinical Data Archive for Head and Neck Squamous
+    Cell Carcinoma Patients Treated with Radiotherapy
+  remarks: Original data at https://www.cancerimagingarchive.net/collection/hnscc/.
+  software: PyRadiomics
+  year: 2018
+Head-Neck-PET-CT:
+  BrokenFeatures: 0
+  ClassBalance: 67
+  Dimensionality: 2.33
+  Missings: 0
+  authors: "Valli\xE8res et al."
+  dataset: Head-Neck-PET-CT
+  download_source: https://radiomics.uk/
+  modality: PET/CT
+  nFeatures: 210
+  nInstances: 91
+  outcome: Lymph node metastasis presence
+  pathology: Head-and-neck cancer
+  publication_doi: https://doi.org/10.1038/s41598-017-10371-5
+  publication_title: Radiomics strategies for risk assessment of tumour failure in
+    head-and-neck cancer
+  remarks: Original data at https://www.cancerimagingarchive.net/collection/head-neck-pet-ct/.
+  software: PyRadiomics
+  year: 2017
+Head-Neck-Radiomics-HN1:
+  BrokenFeatures: 0
+  ClassBalance: 45
+  Dimensionality: 0.78
+  Missings: 0
+  authors: Aerts et al.
+  dataset: Head-Neck-Radiomics-HN1
+  download_source: https://radiomics.uk/
+  modality: CT
+  nFeatures: 105
+  nInstances: 137
+  outcome: Lymph node metastasis presence
+  pathology: Head-and-neck cancer
+  publication_doi: http://doi.org/10.1038/ncomms5006
+  publication_title: Decoding Tumour Phenotype by Noninvasive Imaging Using a Quantitative
+    Radiomics Approach
+  remarks: Original data at https://www.cancerimagingarchive.net/collection/head-neck-radiomics-hn1/
+  software: PyRadiomics
+  year: 2014
+Hosny2018A:
+  BrokenFeatures: 1
+  ClassBalance: 54
+  Dimensionality: 3.37
+  Missings: 2258
+  authors: Hosny et al.
+  dataset: Hosny2018A
+  download_source: https://github.com/modelhub-ai/deep-prognosis
+  modality: CT
+  nFeatures: 984
+  nInstances: 293
+  outcome: Overall survival (at 2 years)
+  pathology: NSCLC
+  publication_doi: https://doi.org/10.1371/journal.pmed.1002711
+  publication_title: 'Deep learning for lung cancer prognostication: A retrospective
+    multi-cohort radiomics study'
+  remarks: HarvardRT subset. Removed exponential_glcm features since they exhibited
+    NA values due to numerical overflow.
+  software: PyRadiomics
+  year: 2018
+Hosny2018B:
+  BrokenFeatures: 1
+  ClassBalance: 29
+  Dimensionality: 4.76
+  Missings: 1157
+  authors: Hosny et al.
+  dataset: Hosny2018B
+  download_source: https://github.com/modelhub-ai/deep-prognosis
+  modality: CT
+  nFeatures: 984
+  nInstances: 207
+  outcome: Overall survival (at 2 years)
+  pathology: NSCLC
+  publication_doi: https://doi.org/10.1371/journal.pmed.1002711
+  publication_title: 'Deep learning for lung cancer prognostication: A retrospective
+    multi-cohort radiomics study'
+  remarks: Maastro subset. Removed exponential_glcm features since they exhibited
+    NA values due to numerical overflow.
+  software: PyRadiomics
+  year: 2018
+Hosny2018C:
+  BrokenFeatures: 1
+  ClassBalance: 73
+  Dimensionality: 5.39
+  Missings: 169
+  authors: Hosny et al.
+  dataset: Hosny2018C
+  download_source: https://github.com/modelhub-ai/deep-prognosis
+  modality: CT
+  nFeatures: 984
+  nInstances: 183
+  outcome: Overall survival (at 2 years)
+  pathology: NSCLC
+  publication_doi: https://doi.org/10.1371/journal.pmed.1002711
+  publication_title: 'Deep learning for lung cancer prognostication: A retrospective
+    multi-cohort radiomics study'
+  remarks: Moffitt subset. Removed exponential_glcm features since they exhibited
+    NA values due to numerical overflow.
+  software: PyRadiomics
+  year: 2018
+Huang2023:
+  BrokenFeatures: 0
+  ClassBalance: 46
+  Dimensionality: 4.04
+  Missings: 0
+  authors: Huang et al.
+  dataset: Huang2023
+  download_source: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0292110#sec019
+  modality: CT
+  nFeatures: 855
+  nInstances: 212
+  outcome: Malignancy presence
+  pathology: Renal cancer
+  publication_doi: https://doi.org/10.1371/journal.pone.0292110
+  publication_title: 'Enhanced and unenhanced: Radiomics models for discriminating
+    between benign and malignant cystic renal masses on CT images: A multi-center
+    study'
+  remarks: Unenhanced CT features were used
+  software: PyRadiomics
+  year: 2023
+Hunter2023:
+  BrokenFeatures: 0
+  ClassBalance: 54
+  Dimensionality: 3.85
+  Missings: 0
+  authors: Hunter et al.
+  dataset: Hunter2023
+  download_source: https://data.mendeley.com/datasets/rxn95mp24d/1
+  modality: CT
+  nFeatures: 1998
+  nInstances: 520
+  outcome: Nodule malignancy presence
+  pathology: Lung cancer
+  publication_doi: https://doi.org/10.1038/s41416-023-02480-y
+  publication_title: Radiomics-based decision support tool assists radiologists in
+    small lung nodule classification and improves lung cancer early diagnosis
+  remarks: null
+  software: TexLab
+  year: 2023
+ISPY1:
+  BrokenFeatures: 9
+  ClassBalance: 57
+  Dimensionality: 2.31
+  Missings: 0
+  authors: Newitt et al.
+  dataset: ISPY1
+  download_source: https://www.cancerimagingarchive.net/analysis-result/ispy1-tumor-seg-radiomics/
+  modality: MRI
+  nFeatures: 370
+  nInstances: 161
+  outcome: Hormone receptor positive status
+  pathology: Breast cancer
+  publication_doi: http://doi.org/10.7937/K9/TCIA.2016.HdHpgJLK
+  publication_title: Multi-center breast DCE-MRI data and segmentations from patients
+    in the I-SPY 1/ACRIN 6657 trials
+  remarks: TCIA sources at https://www.cancerimagingarchive.net/collection/ispy1/,
+    features from the cited analysis were used.
+  software: CaPTk
+  year: 2016
+Keek2020:
+  BrokenFeatures: 10
+  ClassBalance: 44
+  Dimensionality: 4.85
+  Missings: 52
+  authors: Keek et al.
+  dataset: Keek2020
+  download_source: https://github.com/SebastianSanduleanu/Peritumoral-HN-Radiomics
+  modality: CT
+  nFeatures: 1322
+  nInstances: 273
+  outcome: Overall survival (at 3 years)
+  pathology: HNSCC
+  publication_doi: https://doi.org/10.1371/journal.pone.0232639
+  publication_title: Computed tomography-derived radiomic signature of head and neck
+    squamous cell carcinoma (peri)tumoral tissue for the prediction of locoregional
+    recurrence and distant metastasis after concurrent chemo-radiotherapy
+  remarks: Due to incompatibilities, only the DESIGN data was used. Patients with
+    less than 3 years of follow-up were removed.
+  software: Inhouse
+  year: 2020
+LGG-1p19qDeletion:
+  BrokenFeatures: 0
+  ClassBalance: 64
+  Dimensionality: 12.78
+  Missings: 0
+  authors: Akkus et al.
+  dataset: LGG-1p19qDeletion
+  download_source: https://radiomics.uk/
+  modality: MRI
+  nFeatures: 2030
+  nInstances: 159
+  outcome: 1p19q co-deletion status
+  pathology: Glioma
+  publication_doi: https://doi.org/10.1007/s10278-017-9984-3
+  publication_title: Predicting Deletion of Chromosomal Arms 1p/19q in Low-Grade Gliomas
+    from MR Images Using Machine Intelligence
+  remarks: Original data at https://www.cancerimagingarchive.net/collection/lgg-1p19qdeletion/
+  software: PyRadiomics
+  year: 2017
+LNDb:
+  BrokenFeatures: 0
+  ClassBalance: 66
+  Dimensionality: 0.62
+  Missings: 1
+  authors: Pedrosa et al.
+  dataset: LNDb
+  download_source: https://radiomics.uk/
+  modality: CT
+  nFeatures: 105
+  nInstances: 173
+  outcome: Fleischner score group
+  pathology: Lung cancer
+  publication_doi: https://doi.org/10.48550/arXiv.1911.08434
+  publication_title: 'LNDb: a lung nodule database on computed tomography'
+  remarks: Original dataset et https://zenodo.org/records/6613714#.Y8vLZezMIUq
+  software: PyRadiomics
+  year: 2019
+Li2020:
+  BrokenFeatures: 0
+  ClassBalance: 63
+  Dimensionality: 7.8
+  Missings: 0
+  authors: Li et al.
+  dataset: Li2020
+  download_source: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0227703#sec017
+  modality: MRI
+  nFeatures: 396
+  nInstances: 51
+  outcome: Immunohistochemical result
+  pathology: Glioma
+  publication_doi: https://doi.org/10.1371/journal.pone.0227703
+  publication_title: 'High-order radiomics features based on T2 FLAIR MRI predict
+    multiple glioma immunohistochemical features: A more precise and personalized
+    gliomas management'
+  remarks: Outcome is based on Ki-67, S-100, vimentin and CD34 immunohistochemical
+    results (called Label in the study)
+  software: Artificial Intelligence Kit (Commercial)
+  year: 2020
+Lu2019:
+  BrokenFeatures: 0
+  ClassBalance: 73
+  Dimensionality: 8.79
+  Missings: 0
+  authors: Lu et al.
+  dataset: Lu2019
+  download_source: https://doi.org/10.17632/4c5znk5m8t.1
+  modality: CT
+  nFeatures: 657
+  nInstances: 75
+  outcome: Progression free survival (at 2 years)
+  pathology: Ovarian cancer
+  publication_doi: https://doi.org/10.1038/s41467-019-08718-9
+  publication_title: A mathematical-descriptor of tumor-mesoscopic-structure from
+    computed-tomography images annotates prognostic- and molecular-phenotypes of epithelial
+    ovarian cancer
+  remarks: Only patients with follow-up at least 2 years were included, only those
+    with bilateral tumors (Code > 1).
+  software: Inhouse
+  year: 2019
+Meningioma-SEG-CLASS:
+  BrokenFeatures: 0
+  ClassBalance: 43
+  Dimensionality: 23.09
+  Missings: 0
+  authors: Vassantachart et al.
+  dataset: Meningioma-SEG-CLASS
+  download_source: https://radiomics.uk/
+  modality: MRI
+  nFeatures: 2030
+  nInstances: 88
+  outcome: Tumor grade
+  pathology: Meningioma
+  publication_doi: https://doi.org/10.1038/s41598-022-07859-0
+  publication_title: Automatic differentiation of Grade I and II meningiomas on magnetic
+    resonance image using an asymmetric convolutional neural network
+  remarks: Original dataset at https://www.cancerimagingarchive.net/collection/meningioma-seg-class/
+  software: PyRadiomics
+  year: 2022
+NSCLC-Radiogenomics:
+  BrokenFeatures: 0
+  ClassBalance: 16
+  Dimensionality: 0.74
+  Missings: 0
+  authors: Gevaert et al.
+  dataset: NSCLC-Radiogenomics
+  download_source: https://radiomics.uk/
+  modality: PET/CT
+  nFeatures: 105
+  nInstances: 144
+  outcome: EGFR mutation status
+  pathology: NSCLC
+  publication_doi: http://doi.org/10.1148/radiol.12111607
+  publication_title: "Non\u2013Small Cell Lung Cancer: Identifying Prognostic Imaging\
+    \ Biomarkers by Leveraging Public Gene Expression Microarray Data\u2014Methods\
+    \ and Preliminary Results"
+  remarks: Original dataset at https://www.cancerimagingarchive.net/collection/nsclc-radiogenomics/
+  software: PyRadiomics
+  year: 2012
+OcanaTienda2023:
+  BrokenFeatures: 0
+  ClassBalance: 48
+  Dimensionality: 16.9
+  Missings: 0
+  authors: "Oca\xF1a-Tienda et al."
+  dataset: OcanaTienda2023
+  download_source: https://springernature.figshare.com/articles/dataset/Radiomic_data_of_annotated_brain_metastases/22277638/1
+  modality: MRI
+  nFeatures: 1130
+  nInstances: 67
+  outcome: Survival (at 2 years)
+  pathology: Brain metastasis
+  publication_doi: https://doi.org/10.1038/s41597-023-02123-0
+  publication_title: A comprehensive dataset of annotated brain metastasis MR images
+    with clinical and radiomic data
+  remarks: All patients died, therefore all data was processed.
+  software: PyRadiomics
+  year: 2023
+PI-CAI:
+  BrokenFeatures: 0
+  ClassBalance: 66
+  Dimensionality: 3.14
+  Missings: 0
+  authors: Saha et al.
+  dataset: PI-CAI
+  download_source: https://radiomics.uk/
+  modality: MRI
+  nFeatures: 3045
+  nInstances: 969
+  outcome: Gleason score risk group
+  pathology: Prostate cancer
+  publication_doi: https://doi.org/10.1016/j.media.2021.102155
+  publication_title: 'End-to-end prostate cancer detection in bpMRI via 3D CNNs: Effects
+    of attention mechanisms, clinical priori and decoupled false positive reduction'
+  remarks: Original dataset at https://pi-cai.grand-challenge.org/
+  software: PyRadiomics
+  year: 2021
+Petrillo2023:
+  BrokenFeatures: 0
+  ClassBalance: 37
+  Dimensionality: 6.66
+  Missings: 0
+  authors: Petrillo et al.
+  dataset: Petrillo2023
+  download_source: https://zenodo.org/records/8392800
+  modality: MRI
+  nFeatures: 851
+  nInstances: 128
+  outcome: Luminal type presence
+  pathology: Breast cancer
+  publication_doi: https://doi.org/10.1007/s11547-023-01718-2
+  publication_title: Radiomics and artificial intelligence analysis by T2-weighted
+    imaging and dynamic contrast-enhanced magnetic resonance imaging to predict Breast
+    Cancer Histological Outcome
+  remarks: Since luminal vs non-luminal had highest AUC, this was selected as endpoint.
+    All features (T2, DCE) were merged.
+  software: PyRadimomics
+  year: 2023
+Prostate-MRI-US-Biopsy:
+  BrokenFeatures: 0
+  ClassBalance: 77
+  Dimensionality: 1.32
+  Missings: 0
+  authors: Sonn et al.
+  dataset: Prostate-MRI-US-Biopsy
+  download_source: https://radiomics.uk/
+  modality: MRI
+  nFeatures: 1015
+  nInstances: 773
+  outcome: Gleason score risk group
+  pathology: Prostate cancer
+  publication_doi: https://doi.org/10.1016/j.juro.2012.08.095
+  publication_title: Targeted biopsy in the detection of prostate cancer using an
+    office based magnetic resonance ultrasound fusion device
+  remarks: Original dataset at https://www.cancerimagingarchive.net/collection/prostate-mri-us-biopsy/
+  software: PyRadiomics
+  year: 2013
+QIN-HEADNECK:
+  BrokenFeatures: 0
+  ClassBalance: 75
+  Dimensionality: 3.59
+  Missings: 0
+  authors: Fedorov et al.
+  dataset: QIN-HEADNECK
+  download_source: https://radiomics.uk/
+  modality: PET/CT
+  nFeatures: 210
+  nInstances: 59
+  outcome: Lymph node metastasis presence
+  pathology: Head-and-neck cancer
+  publication_doi: https://doi.org/10.7717/peerj.2057
+  publication_title: 'DICOM for quantitative imaging biomarker development: a standards
+    based approach to sharing clinical data and structured PET/CT analysis results
+    in head and neck cancer research.'
+  remarks: Original dataset at https://www.cancerimagingarchive.net/collection/qin-headneck/.
+    CT features were used.
+  software: PyRadiomics
+  year: 2016
+Ramella2018:
+  BrokenFeatures: 2
+  ClassBalance: 55
+  Dimensionality: 2.68
+  Missings: 0
+  authors: Ramella et al.
+  dataset: Ramella2018
+  download_source: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0207455#sec010
+  modality: CT
+  nFeatures: 242
+  nInstances: 91
+  outcome: Adaptive chemoradiation presence
+  pathology: NSCLC
+  publication_doi: https://doi.org/10.1371/journal.pone.0207455
+  publication_title: A radiomic approach for adaptive radiotherapy in non-small cell
+    lung cancer patients
+  remarks: null
+  software: Inhouse
+  year: 2018
+Sasaki2019:
+  BrokenFeatures: 0
+  ClassBalance: 49
+  Dimensionality: 4.27
+  Missings: 0
+  authors: Sasaki et al.
+  dataset: Sasaki2019
+  download_source: https://www.nature.com/articles/s41598-019-50849-y#Sec11
+  modality: MRI
+  nFeatures: 587
+  nInstances: 138
+  outcome: MGMT mutation status
+  pathology: Glioblastoma
+  publication_doi: https://doi.org/10.1038/s41598-019-50849-y
+  publication_title: Radiomics and MGMT promoter methylation for prognostication of
+    newly diagnosed glioblastoma
+  remarks: null
+  software: Inhouse
+  year: 2019
+Song2020:
+  BrokenFeatures: 0
+  ClassBalance: 49
+  Dimensionality: 1.02
+  Missings: 0
+  authors: Song et al.
+  dataset: Song2020
+  download_source: https://github.com/salan668/FAE/tree/master/Feature
+  modality: MRI
+  nFeatures: 264
+  nInstances: 260
+  outcome: Clinical significance
+  pathology: Prostate cancer
+  publication_doi: https://doi.org/10.1371/journal.pone.0237587
+  publication_title: 'FeAture Explorer (FAE): A tool for developing and comparing
+    radiomics models'
+  remarks: null
+  software: PyRadiomics
+  year: 2020
+UCSF-PDGM:
+  BrokenFeatures: 0
+  ClassBalance: 89
+  Dimensionality: 17.0
+  Missings: 77
+  authors: Calabrese et al.
+  dataset: UCSF-PDGM
+  download_source: https://radiomics.uk/
+  modality: MRI
+  nFeatures: 7105
+  nInstances: 418
+  outcome: IDH mutation status
+  pathology: Brain cancer
+  publication_doi: https://doi.org/10.1148/ryai.220058
+  publication_title: The UCSF Preoperative Diffuse Glioma MRI (UCSF-PDGM) Dataset
+  remarks: Original dataset at https://www.cancerimagingarchive.net/collection/ucsf-pdgm/.
+    Features of the enhancing tumor were useed.
+  software: PyRadiomics
+  year: 2022
+UPENN-GBM:
+  BrokenFeatures: 0
+  ClassBalance: 42
+  Dimensionality: 59.72
+  Missings: 0
+  authors: Bakas et al.
+  dataset: UPENN-GBM
+  download_source: https://radiomics.uk/
+  modality: MRI
+  nFeatures: 11165
+  nInstances: 187
+  outcome: MGMT mutation status
+  pathology: Glioblastoma
+  publication_doi: https://doi.org/10.1038/s41597-022-01560-7
+  publication_title: 'The University of Pennsylvania glioblastoma (UPenn-GBM) cohort:
+    advanced MRI, clinical, genomics, & radiomics'
+  remarks: Original data at https://www.cancerimagingarchive.net/collection/upenn-gbm/.  Features
+    of the enhancing tumor were useed.
+  software: PyRadiomics
+  year: 2022
+Veeraraghavan2020:
+  BrokenFeatures: 2
+  ClassBalance: 31
+  Dimensionality: 1.35
+  Missings: 0
+  authors: Veeraraghavan et al.
+  dataset: Veeraraghavan2020
+  download_source: https://github.com/The-Veeraraghavan-Lab/SciRepEndometrial2020
+  modality: MRI
+  nFeatures: 200
+  nInstances: 150
+  outcome: High tumor burden presence
+  pathology: Breast cancer
+  publication_doi: https://doi.org/10.1038/s41598-020-72475-9
+  publication_title: Machine learning-based prediction of microsatellite instability
+    and high tumor mutation burden from contrast-enhanced computed tomography in endometrial
+    cancers
+  remarks: null
+  software: CERR
+  year: 2020
+WORC-CRLM:
+  BrokenFeatures: 0
+  ClassBalance: 48
+  Dimensionality: 13.21
+  Missings: 0
+  authors: Starmans et. al.
+  dataset: WORC-CRLM
+  download_source: https://radiomics.uk/
+  modality: CT
+  nFeatures: 1015
+  nInstances: 77
+  outcome: Histopathological growth pattern
+  pathology: Colorectal cancer
+  publication_doi: https://doi.org/10.48550/arXiv.2108.08618
+  publication_title: Reproducible radiomics through automated machine learning validated
+    on twelve clinical applications
+  remarks: Original data at https://xnat.bmia.nl/data/projects/worc
+  software: PyRadiomics
+  year: 2021
+WORC-Desmoid:
+  BrokenFeatures: 0
+  ClassBalance: 35
+  Dimensionality: 5.01
+  Missings: 0
+  authors: Starmans et. al.
+  dataset: WORC-Desmoid
+  download_source: https://radiomics.uk/
+  modality: MRI
+  nFeatures: 1015
+  nInstances: 203
+  outcome: Fibromatosis presence
+  pathology: Soft-tissue sarcoma
+  publication_doi: https://doi.org/10.48550/arXiv.2108.08618
+  publication_title: Reproducible radiomics through automated machine learning validated
+    on twelve clinical applications
+  remarks: Original data at https://xnat.bmia.nl/data/projects/worc
+  software: PyRadiomics
+  year: 2021
+WORC-GIST:
+  BrokenFeatures: 0
+  ClassBalance: 51
+  Dimensionality: 4.15
+  Missings: 0
+  authors: Starmans et. al.
+  dataset: WORC-GIST
+  download_source: https://radiomics.uk/
+  modality: CT
+  nFeatures: 1015
+  nInstances: 245
+  outcome: Gastrointestinal stromal tumors presence
+  pathology: Gastrointestinal cancer
+  publication_doi: https://doi.org/10.48550/arXiv.2108.08618
+  publication_title: Reproducible radiomics through automated machine learning validated
+    on twelve clinical applications
+  remarks: Original data at https://xnat.bmia.nl/data/projects/worc
+  software: PyRadiomics
+  year: 2021
+WORC-Lipo:
+  BrokenFeatures: 0
+  ClassBalance: 50
+  Dimensionality: 8.92
+  Missings: 0
+  authors: Starmans et. al.
+  dataset: WORC-Lipo
+  download_source: https://radiomics.uk/
+  modality: MRI
+  nFeatures: 1015
+  nInstances: 114
+  outcome: Malignancy presence
+  pathology: Liposarcoma
+  publication_doi: https://doi.org/10.48550/arXiv.2108.08618
+  publication_title: Reproducible radiomics through automated machine learning validated
+    on twelve clinical applications
+  remarks: Original data at https://xnat.bmia.nl/data/projects/worc
+  software: PyRadiomics
+  year: 2021
+WORC-Liver:
+  BrokenFeatures: 0
+  ClassBalance: 51
+  Dimensionality: 5.47
+  Missings: 0
+  authors: Starmans et. al.
+  dataset: WORC-Liver
+  download_source: https://radiomics.uk/
+  modality: MRI
+  nFeatures: 1015
+  nInstances: 186
+  outcome: Malignancy presence
+  pathology: Liver cancer
+  publication_doi: https://doi.org/10.48550/arXiv.2108.08618
+  publication_title: Reproducible radiomics through automated machine learning validated
+    on twelve clinical applications
+  remarks: Original data at https://xnat.bmia.nl/data/projects/worc
+  software: PyRadiomics
+  year: 2021
+WORC-Melanoma:
+  BrokenFeatures: 0
+  ClassBalance: 49
+  Dimensionality: 10.71
+  Missings: 0
+  authors: Starmans et. al.
+  dataset: WORC-Melanoma
+  download_source: https://radiomics.uk/
+  modality: CT
+  nFeatures: 1015
+  nInstances: 95
+  outcome: BRAF mutation status
+  pathology: Melanoma
+  publication_doi: https://doi.org/10.48550/arXiv.2108.08618
+  publication_title: Reproducible radiomics through automated machine learning validated
+    on twelve clinical applications
+  remarks: Original data at https://xnat.bmia.nl/data/projects/worc
+  software: PyRadiomics
+  year: 2021
+Wang2024:
+  BrokenFeatures: 0
+  ClassBalance: 40
+  Dimensionality: 4.21
+  Missings: 336
+  authors: Wang et al.
+  dataset: Wang2024
+  download_source: https://doi.org/10.6084/m9.figshare.25203245
+  modality: MRI
+  nFeatures: 280
+  nInstances: 67
+  outcome: EGFR mutation status
+  pathology: Glioblastoma
+  publication_doi: https://doi.org/10.1371/journal.pone.0299267
+  publication_title: Quantifying intra-tumoral genetic heterogeneity of glioblastoma
+    toward precision medicine using MRI and a data-inclusive machine learning algorithm
+  remarks: Tumoral ROIs were used. Sliding features were averaged. Multiple biopsies
+    were ignored by taking the first. Three samples were removed since the T2 was
+    missing in these.
+  software: Inhouse
+  year: 2024
+Zhang2023:
+  BrokenFeatures: 0
+  ClassBalance: 51
+  Dimensionality: 8.78
+  Missings: 0
+  authors: Zhang et al.
+  dataset: Zhang2023
+  download_source: https://peerj.com/articles/14559/#supplemental-information
+  modality: CT
+  nFeatures: 1781
+  nInstances: 203
+  outcome: Histological invasiveness
+  pathology: Lung cancer
+  publication_doi: https://doi.org/10.7717/peerj.14559
+  publication_title: 'Computed tomography-based radiomics machine learning models
+    for prediction of histological invasiveness with sub-centimeter subsolid pulmonary
+    nodules: a retrospective study'
+  remarks: Lesions, not patient-wise data.
+  software: PyRadiomics
+  year: 2023
+Zhang2024A:
+  BrokenFeatures: 0
+  ClassBalance: 57
+  Dimensionality: 15.11
+  Missings: 0
+  authors: Zhang et al.
+  dataset: Zhang2024A
+  download_source: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0300170#sec017
+  modality: PET/CT
+  nFeatures: 3850
+  nInstances: 255
+  outcome: Histological type
+  pathology: Lung cancer
+  publication_doi: https://doi.org/10.1371/journal.pone.0300170
+  publication_title: Machine learning for differentiating lung squamous cell cancer
+    from adenocarcinoma using Clinical-Metabolic characteristics and 18F-FDG PET/CT
+    radiomics
+  remarks: null
+  software: PyRadiomics
+  year: 2024
+Zhang2024B:
+  BrokenFeatures: 0
+  ClassBalance: 66
+  Dimensionality: 4.35
+  Missings: 0
+  authors: Zhang et al.
+  dataset: Zhang2024B
+  download_source: https://peerj.com/articles/17111/#supplemental-information
+  modality: CT
+  nFeatures: 833
+  nInstances: 192
+  outcome: Lymph node metastasis presence
+  pathology: Gastric cancer
+  publication_doi: https://doi.org/10.7717/peerj.17111
+  publication_title: 'Predictive nomogram for lymph node metastasis and survival in
+    gastric cancer using contrast-enhanced computed tomography-based radiomics: a
+    retrospective study'
+  remarks: null
+  software: PyRadiomics
+  year: 2024

radMLBench/radMLBench.py

@@ -0,0 +1,214 @@
+# -*- coding: utf-8 -*-
+
+"""
+radMLBench was developed at the University Hospital in Essen, Germany.
+For questions use github issues or write an email (aydin.demircioglu@uk-essen.de).
+
+radMLBench is partially based on the PMLB (Penn Machine Learning Benchmarks), see https://epistasislab.github.io/pmlb/
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of this software
+and associated documentation files (the "Software"), to deal in the Software without restriction,
+including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense,
+and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so,
+subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all copies or substantial
+portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT
+LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+"""
+
+import pandas as pd
+import numpy as np
+import os
+from pkg_resources import resource_filename
+import requests
+import warnings
+import subprocess
+import pathlib
+import yaml
+
+from sklearn.model_selection import RepeatedStratifiedKFold
+from sklearn.preprocessing import StandardScaler
+import numpy as np
+
+GITHUB_URL = 'https://github.com/aydindemircioglu/radMLBench/raw/main/datasets'
+suffix = '.gz'
+metaData = None
+
+
+
+def listDatasets (sort_by = None):
+    """Return a list of all available datasets
+
+    Parameters
+    ----------
+    sort_by: str
+        The column to sort by. If column does not exist, the alphabetically sorted list will be returned.
+
+    Returns
+    ----------
+    datasets: list
+        The list of all available datasets, sorted by sort_by.
+    """
+    # ensure its loaded
+    global metaData
+    readMetaData()
+    dsets = list(metaData.keys())
+    try:
+        if sort_by is not None:
+            dsets = sorted(metaData.keys(), key=lambda x: metaData[x][sort_by])
+    except:
+        pass
+    return dsets
+
+
+
+def get_dataset_url(GITHUB_URL, dataset_name, suffix):
+    dataset_url = '{GITHUB_URL}/{DATASET_NAME}{SUFFIX}'.format(
+                                GITHUB_URL=GITHUB_URL,
+                                DATASET_NAME=dataset_name,
+                                SUFFIX=suffix
+                                )
+
+    re = requests.get(dataset_url)
+    if re.status_code != 200:
+        raise ValueError(f'Unable to retrieve dataset from {dataset_url}')
+    return dataset_url
+
+
+
+def loadData (dataset_name,return_X_y=False, local_cache_dir=None):
+    """Download a data set from the radMLBench, (optionally) store it locally, and return the data set.
+
+    You must be connected to the internet if you are fetching a data set that is not cached locally.
+
+    Parameters
+    ----------
+    dataset_name: str
+        The name of the data set to load from PMLB.
+    return_X_y: bool (default: False)
+        Whether to return the data in scikit-learn format, with the features
+        and labels stored in separate NumPy arrays.
+    local_cache_dir: str (default: None)
+        The directory on your local machine to store the data files.
+        If None, then the local data cache will not be used.
+
+    Returns
+    ----------
+    dataset: pd.DataFrame or (array-like, array-like)
+        if return_X_y == False: A pandas DataFrame containing the fetched data set (with columns ID and Target)
+        if return_X_y == True: A tuple of NumPy arrays containing (features, labels)
+    """
+
+    if dataset_name not in listDatasets():
+        raise ValueError('Dataset not found?')
+
+    if local_cache_dir is None:
+        dataset_url = get_dataset_url(GITHUB_URL, dataset_name, suffix)
+        dataset = pd.read_csv(dataset_url, compression='gzip')
+    else:
+        dataset_path = os.path.join(local_cache_dir, dataset_name+suffix)
+        if os.path.exists(dataset_path):
+            dataset = pd.read_csv(dataset_path, compression='gzip')
+        else:
+            print (f"Downloading to {dataset_path}...")
+            dataset_url = get_dataset_url(GITHUB_URL, dataset_name, suffix)
+            dataset = pd.read_csv(dataset_url, compression='gzip')
+            dataset_dir = os.path.split(dataset_path)[0]
+            if not os.path.isdir(dataset_dir):
+                os.makedirs(dataset_dir)
+            dataset.to_csv(dataset_path, compression='gzip', index=False)
+
+    if return_X_y:
+        X = dataset.drop(['ID', 'Target'], axis=1).values
+        y = dataset['Target'].values
+        return (X, y)
+    else:
+        return dataset
+
+
+
+def getCVSplits(dataset, num_splits=10, num_repeats=10):
+    """
+    Generate indices for cross-validation splits.
+
+    Parameters:
+    - dataset (str or DataFrame or tuple of numpy arrays): Input dataset. If str, it will be treated as dataset name to load.
+    - num_splits (int): Number of folds for cross-validation.
+    - num_repeats (int): Number of times to repeat the cross-validation process.
+
+    Returns:
+    - List of tuples: Each tuple contains train and test indices for one split.
+    """
+    if isinstance(dataset, str):
+        dataset = loadDataset(dataset)
+
+    random_state = 42 * num_splits + num_repeats + 42
+
+    if isinstance(dataset, pd.DataFrame):
+        y = dataset["Target"].values
+        X = dataset.drop(columns=["Target", "ID"]).values
+    else:
+        X, y = dataset
+
+    cv_splits = []
+    rskf = RepeatedStratifiedKFold(n_splits=num_splits, n_repeats=num_repeats, random_state=random_state)
+    for train_index, test_index in rskf.split(X, y):
+        cv_splits.append((train_index, test_index))
+
+    return cv_splits
+
+
+
+def readMetaData ():
+    """Internal. Reads metadata into variable"""
+    def read_yaml(file_path):
+        try:
+            with open(file_path, 'r') as stream:
+                data = yaml.safe_load(stream)
+                return data
+        except Exception as e:
+            print ('Unable to read metadata file for all dataset!')
+            print ('Error', e)
+            return None
+
+    global metaData
+    if metaData is None:
+        #print ("Reading metadata.")
+        package_dir = resource_filename('radMLBench', '')
+        metadata_path = os.path.join(package_dir, 'metadata.yaml')
+        metaData = read_yaml(metadata_path)
+    pass
+
+
+
+
+def getMetaData (dataset_name):
+    """Retrieve the metadata for a data set from the radMLBench.
+
+    Parameters
+    ----------
+    dataset_name: str
+        The name of the data set to retrieve the metadata for.
+
+    Returns
+    ----------
+    metadata: dictionary
+        Dictionary containing the metadata
+    """
+    # ensure its loaded
+    global metaData
+    readMetaData()
+    if dataset_name not in metaData:
+        print ("Unknown dataset.")
+        return None
+    return metaData[dataset_name]
+
+
+
+#

radMLBench-1.0.dist-info/LICENSE.txt

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024, Aydin Demircioglu
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

radMLBench-1.0.dist-info/METADATA

@@ -0,0 +1,31 @@
+Metadata-Version: 2.1
+Name: radMLBench
+Version: 1.0
+Summary: A Python wrapper for the radMLBench data repository.
+Home-page: https://github.com/aydindemircioglu/radMLBench
+Author: Aydin Demircioglu
+Author-email: aydin.demircioglu@uk-essen.de
+License: MIT
+Keywords: radiomics,data mining,benchmark,machine learning,data analysis,data sets,data science,wrapper
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3.7
+Requires-Python: >=3.6.0
+Description-Content-Type: text/markdown
+License-File: LICENSE.txt
+Requires-Dist: pandas (>=1.2.0)
+Requires-Dist: requests (>=2.18.0)
+Requires-Dist: pyyaml (>=5.2)
+Requires-Dist: joblib (>=0.12.0)
+Requires-Dist: numpy (>=1.18.0)
+Requires-Dist: scikit-learn (>=0.20.0)
+
+
+A Python wrapper for the radMLBench data repository.
+
+Contact
+=============
+If you have any questions or comments about radMLBench,
+please feel free to contact us via e-mail: aydin.demircioglu@uk-essen.de
+
+This project is hosted at https://github.com/aydindemircioglu/radMLBench

radMLBench-1.0.dist-info/RECORD

@@ -0,0 +1,9 @@
+radMLBench/__init__.py,sha256=yqNSt_MuyFATNDFfO_TQ-Jvb_BIQheRg4M7188y7Q-A,1428
+radMLBench/metadata.yaml,sha256=c5-eKbcXZV4-sHHyIhwmbqSqG0p5DbzcwxmvG_zJPqc,32004
+radMLBench/radMLBench.py,sha256=H9iw7f-38J2se0zA36rUOvvFjWlErQg4rzThYkrKJuM,7127
+radMLBench-1.0.dist-info/LICENSE.txt,sha256=gTSULos27L9Ns-K-6CiVNHN6JNSQKEejyUh5fmuq_E8,1075
+radMLBench-1.0.dist-info/METADATA,sha256=pJYZyOUfqQ3NbWD1p8YXFEdTzLQ5oQM-vc9e1vY5muA,1083
+radMLBench-1.0.dist-info/WHEEL,sha256=pkctZYzUS4AYVn6dJ-7367OJZivF2e8RA9b_ZBjif18,92
+radMLBench-1.0.dist-info/top_level.txt,sha256=5BKhbStaBdsf74i1jicPCX82ERw2V9Ww2tHjAq7sZ48,11
+radMLBench-1.0.dist-info/zip-safe,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
+radMLBench-1.0.dist-info/RECORD,,

radMLBench-1.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: bdist_wheel (0.40.0)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

radMLBench-1.0.dist-info/top_level.txt

@@ -0,0 +1 @@
+radMLBench

radMLBench-1.0.dist-info/zip-safe

@@ -0,0 +1 @@
+