def __init__(self, *args):
DataParser.__init__(self, *args)
# cleanup subjects
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
# self.data['ID'] = self.data['ID'].apply(lambda x: getID(x))
if self.info is None:
self.info = {'prefix': 'GDN', 'xsitype': 'opex:godin'}
# Replace field headers
self.fields = ['strenuous', 'moderate', 'light', 'total', 'sweat']
fields = ['strenuous', 'moderate', 'light', 'total', 'sweat']
cols = [
'ID', 'Strenuous', 'Moderate', 'Light',
'Totalleisureactivityscore', 'Sweat(1,2,or3)'
]
ncols = ['SubjectID']
renamecols = dict(list(zip(cols, ncols + fields)))
df = self.data.iloc[:, 0:7]
df.dropna(axis=0, how='any', thresh=5,
inplace=True) # remove all empty rows
df.fillna(999, inplace=True) # replace any remaining na with 999
df.rename(columns=renamecols, inplace=True)
df.reindex()
self.data = df
# sort subjects
self.sortSubjects('SubjectID')
print('Data load complete')
def __init__(self, *args, **kwargs):
DataParser.__init__(self, *args, **kwargs)
if self.data is None:
raise ValueError('IPAQ Parser: Data not loaded')
self.type = ''
# if 'type' in kwargs:
# self.type = kwargs.get('type')
# self.fields = self.dbi.getFields(self.type)
# self.info = self.dbi.getInfo(self.type)
# elif self.etype is not None:
# self.type = self.etype
self.data.dropna(axis=1, how='all', inplace=True)
# self.data = self.data.filter(regex="^[^78]")
fields = [
'sitting', 'walking_days', 'walking_time', 'moderate_days',
'moderate_time', 'vigorous_days', 'vigorous_time', 'pa', 'mvpa'
]
self.fields = fields
self.data = self.data[['ID'] + fields]
# self.data.columns = ['ID'] + fields
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
# self.data['ID'] = self.data['ID'].apply(lambda x: getID(x))
self.sortSubjects('ID')
print('Data load complete')
if self.info is None:
self.info = {'prefix': 'IP', 'xsitype': 'opex:ipaq'}
def __init__(self, *args):
DataParser.__init__(self, *args)
fields = join(self.resource_dir, "dexa_fields.xlsx")
#Replace field headers
if access(fields, R_OK):
self.fields = pd.read_excel(fields, header=0, sheetname='dexa_fields')
df_header = pd.read_excel(fields, header=0, sheetname='dexa_header')
self.header = df_header['concatenated'].tolist()
self.data.columns = self.header
print(("Loaded rows=", len(self.data['ID'])))
#extract subject info
df_subj = self.data.iloc[:,0:4]
df_subj['SubjectID'] = df_subj.apply(lambda x: stripspaces(x, 'ID'), axis=1)
#Split data into intervals
self.intervals = {0:'BASELINE', 3:'MIDPOINT',6:'ENDPOINT', 9:'MID-FOLLOW-UP', 12:'FOLLOW-UP'}
self.df = dict()
for i,intval in list(self.intervals.items()):
cols = [c for c in self.header if c.startswith(intval)]
simplecols = []
for col in cols:
cparts = col.split("_")
simplecols.append("_".join(cparts[1:]))
self.df[i] = pd.concat([df_subj,self.data[cols]], axis=1)
self.df[i].columns = df_subj.columns.tolist() + simplecols
#self.df[i].reindex(df_subj.columns.tolist() + simplecols, fill_value='')
if DEBUG:
msg ="Interval=%s data=%d" % (intval, len(self.df[i]))
print(msg)
self.sortSubjects('SubjectID')
else:
raise ValueError("Cannot access fields file: %s" % fields)
def __loadEfficiencydata(self, datafile):
# Load efficiency data from single file
effdata = pd.read_excel(datafile, sheet_name=0, header=1)
effdata.drop(effdata.index[0], inplace=True)
effdata['SubjectID'] = effdata.apply(lambda x: stripspaces(x, 'ID'),
axis=1)
logging.info("Loaded Efficiency: %d", len(effdata))
return effdata
def __init__(self, *args):
DataParser.__init__(self, *args)
# cleanup subjects
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
# self.data['ID'] = self.data['ID'].apply(lambda x: getID(x))
df = self.data
# Replace field headers
self.fields = ['current', 'past', 'total']
columns = ['CurrentResult', 'PastResult', 'TotalResult']
df.rename(columns=dict(list(zip(columns, self.fields))), inplace=True)
self.data = df
self.sortSubjects('ID')
print('Data load complete')
def create_cutoffs(inputdir):
dates = pd.read_excel(join(inputdir, 'Dates.xlsx'))
dates['Subject'] = dates.apply(lambda x: stripspaces(x, 'ID'), axis=1)
dates.set_index('Subject', inplace=True)
intervals = [1, 2, 4, 5]
for i in intervals:
cols = [
c for c in dates.columns.tolist()
if bool(re.search('^{} month'.format(str(i)), c))
]
dates[str(i) + ' month assessment'] = dates[cols].values.max(axis=1)
assessments = ['Assessment B', 'Training start date'] + [
str(i) + ' month assessment' for i in list(range(1, 7)) + [9, 12]
]
dates[assessments].to_csv(join(inputdir, 'dates_cutoff.csv'))
def __init__(self, *args, **kwargs):
DataParser.__init__(self, *args)
#Maybe empty sheet
if self.data.empty or len(self.data.columns) <= 1:
msg = "No data available"
raise ValueError(msg)
# cleanup subjects
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
if self.info is None:
self.info = {'prefix': 'INS', 'xsitype': 'opex:insomnia'}
# Replace field headers
self.fieldmap = {
'q1': 'Q1',
'q2': 'Q2',
'q3': 'Q3',
'q4': 'Q4',
'q5': 'Q5',
'q6': 'Q6',
'q7': 'Q7',
'total': 'TotalScore'
}
cols = [
'ID', self.fieldmap['q1'], self.fieldmap['q2'],
self.fieldmap['q3'], self.fieldmap['q4'], self.fieldmap['q5'],
self.fieldmap['q6'], self.fieldmap['q7'], self.fieldmap['total']
]
# self.fieldmap = {'total': 'TotalScore'}
self.fields = list(self.fieldmap.keys())
ncols = ['SubjectID'] + self.fields
# cols = ['ID', self.fieldmap['total']]
# zeros have been entered when should be blank
self.data[self.fieldmap['total']] = self.data.apply(
lambda x: self.nodatarow(x, self.fieldmap['total']), axis=1)
self.data[cols[1:]] = self.data.apply(
lambda x: self.nodatarow(x, cols[1:]), axis=1)
df = self.data[cols]
df = df.astype(object) # convert to object
df.columns = ncols
df.reindex()
self.data = df
# sort subjects
self.sortSubjects('SubjectID')
print('Data load complete')
def __init__(self, *args):
DataParser.__init__(self, *args)
# cleanup subjects
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
# self.data['ID'] = self.data['ID'].apply(lambda x: getID(x))
if self.info is None:
self.info = {'prefix': 'GDN', 'xsitype': 'opex:paces'}
# Replace field headers
self.fields = ['q' + str(i)
for i in range(1, 9)] + ['total', 'enjoy_percent']
columns = [
'Q1', 'Q2', 'Q3', 'Q4', 'Q5', 'Q6', 'Q7', 'Q8', 'SumTotal',
'%Enjoyment'
]
renamecols = dict(list(zip(columns, self.fields)))
self.data.rename(columns=renamecols, inplace=True)
self.sortSubjects('ID')
print('Data load complete')
def __init__(self, *args):
DataParser.__init__(self, *args)
# Maybe empty sheet
if self.data.empty or len(self.data.columns) <= 1:
msg = "No data available"
raise ValueError(msg)
# cleanup subjects
self.data['ID'] = self.data.apply(lambda x: stripspaces(x, 0), axis=1)
# self.data['ID'] = self.data['ID'].apply(lambda x: getID(x))
if self.info is None:
self.info = {'prefix': 'PSQ', 'xsitype': 'opex:psqi'}
# Replace field headers
self.fields = ['c'+str(i) for i in range(1,8)] + ['total']
ncols = ['SubjectID'] + self.fields
cols = ['ID'] + [c for c in self.data.columns if (isinstance(c,str) or isinstance(c,str)) and c.startswith('Component')] + ['total']
df = self.data[cols]
df.columns = ncols
df.reindex()
self.data = df
# sort subjects
self.sortSubjects('SubjectID')
print('Data load complete')