def processCounties( siteResources = SITE_RESOURCES,
countiesFile = COUNTY_CASES_CSBS_FILE,
siteData = SITE_RESOURCES,
outputFileName = COUNTIES_US_FILE):
print('vucounty - processing US counties for the good of humanity')
inputFileName = os.path.join(siteResources, countiesFile)
with open(inputFileName, 'r') as inputFile:
regions = json.load(inputFile)
dataset = dict()
for region in regions:
state = region[STATE_OR_PROVINCE_KEY]
if state not in dataset:
print('updating %s' % state)
dataset[state] = dict()
county = region['county']
latest = region['latest']
print(' updating %s, %s' % (county, state))
if 'recovered' in latest:
del(latest['recovered'])
dataset[state][county] = latest
dumpJSON(dataset, os.path.join(siteData, outputFileName))
return dataset
def _dumpTimeSeriesAsJSON(timeSeries, target=None):
assert isinstance(timeSeries.index, DatetimeIndex)
_castDatetimeIndexToStr(timeSeries)
result = {
timeSeries.name: timeSeries.to_dict(),
}
if target:
dumpJSON(result, target)
return result
def _main(
siteDataDirectory=SITE_RESOURCES,
outFileName=HOSPITAL_BEDS_FILE_NAME,
nStateLimit=None,
):
postCodes = pd.read_csv(STATE_CODES_PATH)
print('vuhospitals - getting the total hospital beds count per state')
bedCount = _getTotalBedCount(postCodes, nStateLimit=nStateLimit)
dumpJSON(bedCount, resolveFileName(siteDataDirectory, outFileName))
def _dumpPredictionCollectionAsJSON(
predictionsPercentilesTS,
regionName,
predictionsPercentiles,
target,
):
result = {}
for i, (qLow, qHigh) in enumerate(predictionsPercentiles):
tsLow = predictionsPercentilesTS[i][0]
tsHigh = predictionsPercentilesTS[i][1]
_castDatetimeIndexToStr(tsLow)
_castDatetimeIndexToStr(tsHigh)
result[qLow] = tsLow.to_dict()
result[qHigh] = tsHigh.to_dict()
result = {regionName: result}
if target:
dumpJSON(result, target)
return result