from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('acarbose', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('glimepiride', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['Male', 'Female', 'Unknown/Invalid']
fd = CatFeatureDescriptor('gender', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('acetohexamide', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('insulin', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
from sklearn.preprocessing import OrdinalEncoder
categories = [
'[0-10)', '[10-20)', '[20-30)', '[30-40)', '[40-50)', '[50-60)', '[60-70)',
'[70-80)', '[80-90)', '[90-100)'
]
fd = CatFeatureDescriptor(name='age', categories=categories)
fd.steps[1] = [(OrdinalEncoder, {'categories': [[categories]]})]
from ads.utils.feature import CatFeatureDescriptor
# TODO 96000 of Nones
categories = ['>300', 'Norm', '>200']
fd = CatFeatureDescriptor('max_glu_serum', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('tolazamide', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
# TODO 96000 of Nones
categories = ['>7', '>8', 'Norm']
fd = CatFeatureDescriptor('A1Cresult', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
# TODO majority of one class
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('glimepiride-pioglitazone', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Ch']
fd = CatFeatureDescriptor('change', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
# TODO the only 'No' value
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('citoglipton', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
# TODO majority of one class
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('metformin-rosiglitazone', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = [
'25', '1', '3', '6', '2', '5', '11', '7', '10', '4', '14', '18', '8', '13',
'12', '16', '17', '22', '23', '9', '20', '15', '24', '28', '19', '27'
]
fd = CatFeatureDescriptor('discharge_disposition_id', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
# TODO majority of one class
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('glipizide-metformin', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('glyburide', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('miglitol', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['1', '7', '2', '4', '5', '6', '20', '3', '17', '8', '9', '14',
'10', '22', '11', '25', '13']
fd = CatFeatureDescriptor('admission_source_id', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('nateglinide', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
# TODO half of the data are Nones
categories = [
'Pediatrics-Endocrinology', 'InternalMedicine', 'Family/GeneralPractice',
'Cardiology', 'Surgery-General', 'Orthopedics', 'Gastroenterology',
'Surgery-Cardiovascular/Thoracic', 'Nephrology',
'Orthopedics-Reconstructive', 'Psychiatry', 'Emergency/Trauma',
'Pulmonology', 'Surgery-Neuro', 'Obsterics&Gynecology-GynecologicOnco',
'ObstetricsandGynecology', 'Pediatrics', 'Hematology/Oncology',
'Otolaryngology', 'Surgery-Colon&Rectal', 'Pediatrics-CriticalCare',
'Endocrinology', 'Urology', 'Psychiatry-Child/Adolescent',
'Pediatrics-Pulmonology', 'Neurology', 'Anesthesiology-Pediatric',
'Radiology', 'Pediatrics-Hematology-Oncology', 'Psychology', 'Podiatry',
'Gynecology', 'Oncology', 'Pediatrics-Neurology', 'Surgery-Plastic',
'Surgery-Thoracic', 'Surgery-PlasticwithinHeadandNeck', 'Ophthalmology',
'Surgery-Pediatric', 'Pediatrics-EmergencyMedicine',
'PhysicalMedicineandRehabilitation', 'InfectiousDiseases',
'Anesthesiology', 'Rheumatology', 'AllergyandImmunology',
'Surgery-Maxillofacial', 'Pediatrics-InfectiousDiseases',
'Pediatrics-AllergyandImmunology', 'Dentistry', 'Surgeon',
'Surgery-Vascular', 'Osteopath', 'Psychiatry-Addictive',
'Surgery-Cardiovascular', 'PhysicianNotFound', 'Hematology', 'Proctology',
'Obstetrics', 'SurgicalSpecialty', 'Radiologist', 'Pathology',
'Dermatology', 'SportsMedicine', 'Speech', 'Hospitalist',
'OutreachServices', 'Cardiology-Pediatric', 'Perinatology',
'Neurophysiology', 'Endocrinology-Metabolism', 'DCPTEAM', 'Resident'
]
fd = CatFeatureDescriptor('medical_specialty', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('glipizide', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Steady', 'Up', 'Down']
fd = CatFeatureDescriptor('metformin', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
categories = ['No', 'Up', 'Steady', 'Down']
fd = CatFeatureDescriptor('rosiglitazone', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
# TODO half of the data are Nones
categories = [
'MC', 'MD', 'HM', 'UN', 'BC', 'SP', 'CP', 'SI', 'DM', 'CM', 'CH', 'PO',
'WC', 'OT', 'OG', 'MP', 'FR'
]
fd = CatFeatureDescriptor('payer_code', categories=categories)
from ads.utils.feature import CatFeatureDescriptor
from sklearn.preprocessing import OrdinalEncoder
# TODO 98k Nones among 100k
categories = [
'[0-25)', '[25-50)', '[50-75)', '[75-100)', '[100-125)', '[125-150)',
'[150-175)', '[175-200)', '>200'
]
fd = CatFeatureDescriptor(name='weight', categories=categories)
# fd.steps[1] = [(OrdinalEncoder, {'categories': [[categories]]})] # TODO if change steps, unk_val is not appended
