def SMap(pathIn="./",
dataFile="",
dataFrame=None,
pathOut="./",
predictFile="",
lib="",
pred="",
E=0,
Tp=1,
knn=0,
tau=1,
theta=0,
exclusionRadius=0,
columns="",
target="",
smapFile="",
jacobians="",
embedded=False,
verbose=False,
const_pred=False,
showPlot=False):
'''S-Map prediction on path/file.'''
# Establish DF as empty list or Pandas DataFrame for SMap()
if dataFile:
DF = pyBindEDM.DF()
elif isinstance(dataFrame, DataFrame):
if dataFrame.empty:
raise Exception("SMap(): dataFrame is empty.")
DF = pyEDM.AuxFunc.PandasDataFrametoDF(dataFrame)
else:
raise Exception("SMap(): Invalid data input.")
# D is a Python dict from pybind11 < cppEDM SMap:
# { "predictions" : {}, "coefficients" : {} }
D = pyBindEDM.SMap(pathIn, dataFile, DF, pathOut, predictFile, lib, pred,
E, Tp, knn, tau, theta, exclusionRadius, columns,
target, smapFile, jacobians, embedded, const_pred,
verbose)
df_pred = DataFrame(D['predictions']) # Convert to pandas DataFrame
df_coef = DataFrame(D['coefficients']) # Convert to pandas DataFrame
if showPlot:
pyEDM.AuxFunc.PlotObsPred(df_pred, dataFile, E, Tp, False)
pyEDM.AuxFunc.PlotCoeff(df_coef, dataFile, E, Tp)
SMapDict = {'predictions': df_pred, 'coefficients': df_coef}
return SMapDict
def SMap( pathIn = "./",
dataFile = "",
dataFrame = None,
pathOut = "./",
predictFile = "",
lib = "",
pred = "",
E = 0,
Tp = 1,
knn = 0,
tau = -1,
theta = 0,
exclusionRadius = 0,
columns = "",
target = "",
smapFile = "",
jacobians = "",
solver = None,
embedded = False,
verbose = False,
const_pred = False,
showPlot = False,
validLib = [],
generateSteps = 0,
parameterList = False
):
'''S-Map prediction on path/file.'''
# Establish DF as empty list or Pandas DataFrame for SMap()
if dataFile :
DF = pyBindEDM.DF()
elif isinstance( dataFrame, DataFrame ) :
if dataFrame.empty :
raise Exception( "SMap(): dataFrame is empty." )
DF = pyEDM.AuxFunc.PandasDataFrametoDF( dataFrame )
else :
raise Exception( "SMap(): Invalid data input." )
# If lib, pred, columns are not string, but iterable, convert to string
if pyEDM.AuxFunc.NotStringIterable( lib ) :
lib = ' '.join( map( str, lib ) )
if pyEDM.AuxFunc.NotStringIterable( pred ) :
pred = ' '.join( map( str, pred ) )
if pyEDM.AuxFunc.NotStringIterable( columns ) :
columns = ' '.join( map( str, columns ) )
# Validate the solver if one was passed in
if solver :
supportedSolvers = [ 'LinearRegression',
'Ridge', 'Lasso', 'ElasticNet',
'RidgeCV', 'LassoCV', 'ElasticNetCV' ]
if not solver.__class__.__name__ in supportedSolvers :
raise Exception( "SMap(): Invalid solver." )
# D is a Python dict from pybind11 < cppEDM SMap:
# { "predictions" : {}, "coefficients" : {}, ["parameters" : {}] }
D = pyBindEDM.SMap( pathIn,
dataFile,
DF,
pathOut,
predictFile,
lib,
pred,
E,
Tp,
knn,
tau,
theta,
exclusionRadius,
columns,
target,
smapFile,
jacobians,
solver,
embedded,
const_pred,
verbose,
validLib,
generateSteps,
parameterList )
df_pred = DataFrame( D['predictions'] ) # Convert to pandas DataFrame
df_coef = DataFrame( D['coefficients'] ) # Convert to pandas DataFrame
SMapDict = { 'predictions' : df_pred, 'coefficients' : df_coef }
if parameterList :
SMapDict[ 'parameters' ] = D[ 'parameters' ]
if showPlot :
if embedded :
E = len( columns.split() )
pyEDM.AuxFunc.PlotObsPred( df_pred, dataFile, E, Tp, False )
pyEDM.AuxFunc.PlotCoeff ( df_coef, dataFile, E, Tp )
return SMapDict
