def main():
'''Test sklearn.linear_model solvers'''
circle = pyEDM.sampleData['circle']
lmSolvers = {
'SVD': None,
'Ridge': Ridge(alpha=0.05),
'Lasso': Lasso(alpha=0.005),
'ElasticNet': ElasticNet(alpha=0.001, l1_ratio=0.001),
'RidgeCV': RidgeCV(),
'LassoCV': LassoCV(cv=5),
'ElasticNetCV': ElasticNetCV(l1_ratio=[.05, .1, .5, .7, .9, .95, 1],
cv=5)
}
smapResults = {}
for solverName in lmSolvers.keys():
print(solverName)
result = pyEDM.SMap(dataFrame=circle,
lib="1 100",
pred="101 198",
embedded=True,
E=2,
theta=3.14,
columns="x y",
target="x",
showPlot=True,
solver=lmSolvers[solverName])
smapResults[solverName] = result
def main():
'''EDM package examples.
df denotes a Pandas DataFrame.
S and M denote dictionaries of Pandas DataFrames'''
dataFiles = [ "TentMap_rEDM.csv",
"TentMapNoise_rEDM.csv",
"circle.csv",
"block_3sp.csv",
"sardine_anchovy_sst.csv" ]
# Create map of module dataFiles pathnames in Files
Files = {}
for file in dataFiles:
filename = "data/" + file
if pkg_resources.resource_exists( EDM.__name__, filename ):
Files[ file ] = \
pkg_resources.resource_filename( EDM.__name__, filename )
else :
raise Exception( "examples.py: Failed to find data file " + \
file + " in EDM package" )
# Note the path argument is empty "", file path is in Files{}
df = EDM.EmbedDimension( "", Files[ "TentMap_rEDM.csv" ], None, "./", "",
"1 100", "201 500", 10, 1, 1,
"TentMap", "", False, False, 4 )
df = EDM.PredictInterval( "", Files[ "TentMap_rEDM.csv" ], None, "./", "",
"1 100", "201 500", 10, 2, 1,
"TentMap", "", False, False, 4 );
df = EDM.PredictNonlinear( "", Files[ "TentMapNoise_rEDM.csv" ], None,
"./", "", "1 100", "201 500", "", 2, 1, 0, 1,
"TentMap", "", False, False, 4 )
# Tent map simplex : specify multivariable columns embedded = True
df = EDM.Simplex( "", Files[ "block_3sp.csv" ], None, "./", "",
"1 99", "100 198", 3, 1, 0, 1, 0,
"x_t y_t z_t", "x_t", True, False, True, True )
# Tent map simplex : Embed column x_t to E=3, embedded = False
df = EDM.Simplex( "", Files[ "block_3sp.csv" ], None, "./", "",
"1 99", "100 195", 3, 1, 0, 1, 0,
"x_t", "x_t", False, False, True, True )
M = EDM.Multiview( "", Files[ "block_3sp.csv" ], None, "./", "",
"1 100", "101 198", 3, 1, 0, 1,
"x_t y_t z_t", "x_t", 0, 0, False, 4, True )
# S-map circle : specify multivariable columns embedded = True
S = EDM.SMap( "", Files[ "circle.csv" ], None, "./", "",
"1 100", "101 198", 2, 1, 0, 1, 4, 0,
"x y", "x", "", "", True, False, True, True )
df = EDM.CCM( "", Files[ "sardine_anchovy_sst.csv" ], None, "./", "",
3, 0, 0, 1, "anchovy", "np_sst",
"10 70 10", 100, True, False, 0, True, True )
def test_smap2(self):
#--------------------------------------------------------
# block_3sp embedded = True
#--------------------------------------------------------
print("--- S-map block_3sp embedded = True ---")
SM = EDM.SMap("", self.Files["block_3sp.csv"], None, "./", "", "1 99",
"100 198", 3, 1, 0, 1, 2, 0, "x_t y_t z_t", "x_t", "",
"", True, False, True, False)
df = SM['predictions']
dfv = EDM.ReadDataFrame("",
self.Files["Smap_embd_block_3sp_pyEDM.csv"])
S1 = dfv.get('Prediction_t(+1)')
S2 = round(df.get('Predictions'), 4)
self.assertTrue(S1.equals(S2))
def test_smap(self):
#--------------------------------------------------------
# circle test embedded = True
#--------------------------------------------------------
print("--- S-map circle embedded = True ---")
dfc = EDM.ReadDataFrame("", self.Files["circle.csv"])
# Passing Pandas DataFrame input to Smap rather than path/file
SM = EDM.SMap("", "", dfc, "", "", "1 100", "101 198", 2, 1, 0, 1, 4,
0, "x y", "x", "", "", True, False, True, False)
df = SM['predictions']
dfv = EDM.ReadDataFrame("", self.Files["Smap_circle_pyEDM.csv"])
S1 = dfv.get('Prediction_t(+1)')
S2 = round(df.get('Predictions'), 4)
self.assertTrue(S1.equals(S2))
import pandas as pd
import pyEDM
import numpy as np
df = pd.read_csv("twospike_data.csv")
# showt that prediction is good (contains non-1 values)
exclusion_matrix = np.zeros((df.shape[0],df.shape[0]))
for idx in range(70,83):
exclusion_matrix[idx-1, -1+np.array(list(range(20,30)))] = 1
'''simplex_out = pyEDM.Simplex(dataFrame=df, E=3, tau=1, knn=20,
exclusionMatrix=pd.DataFrame( exclusion_matrix ),
lib="1 50", pred="71 81", columns="wave")
print(simplex_out)
'''
simplex_out = pyEDM.SMap(dataFrame=df, E=3, tau=1, knn=20,
exclusionMatrix=pd.DataFrame( exclusion_matrix ),
lib="1 50", pred="71 81", columns="wave")
print(simplex_out)
# construct exlcusion matrix( exclude all non-1 values)
def main():
'''EDM package examples.
df denotes a Pandas DataFrame.
S and M denote dictionaries of Pandas DataFrames'''
dataFiles = [ "TentMap_rEDM.csv",
"TentMapNoise_rEDM.csv",
"circle.csv",
"block_3sp.csv",
"sardine_anchovy_sst.csv" ]
# Create map of module dataFiles pathnames in Files
Files = {}
for file in dataFiles:
filename = "data/" + file
if pkg_resources.resource_exists( EDM.__name__, filename ):
Files[ file ] = \
pkg_resources.resource_filename( EDM.__name__, filename )
else :
raise Exception( "examples.py: Failed to find data file " + \
file + " in EDM package" )
# Note the path argument is empty "", file path is in Files{}
df = EDM.EmbedDimension( pathIn = "", dataFile = Files["TentMap_rEDM.csv"],
lib = "1 100", pred = "201 500", maxE = 10,
Tp = 1, tau = -1, exclusionRadius = 0,
columns = "TentMap", target = "TentMap",
validLib = [], numThreads = 4 )
df = EDM.PredictInterval( pathIn = "", dataFile = Files["TentMap_rEDM.csv"],
lib = "1 100", pred = "201 500", maxTp = 10,
E = 2, tau = -1, exclusionRadius = 0,
columns = "TentMap", target = "TentMap",
validLib = [], numThreads = 4 );
df = EDM.PredictNonlinear( pathIn = "",
dataFile = Files[ "TentMapNoise_rEDM.csv" ],
lib = "1 100", pred = "201 500", E = 2,
Tp = 1, knn = 0, tau = -1,
columns = "TentMap", target = "TentMap",
validLib = [], numThreads = 4 )
# Tent map simplex : specify multivariable columns embedded = True
S = EDM.Simplex( pathIn = "", dataFile = Files[ "block_3sp.csv" ],
lib = "1 99", pred = "100 198", E = 3, Tp = 1,
knn = 0, tau = -1, exclusionRadius = 0,
columns = "x_t y_t z_t", target = "x_t",
embedded = True, showPlot = True,
validLib = [], generateSteps = 0, parameterList = False )
# Tent map simplex : Embed column x_t to E=3, embedded = False
S = EDM.Simplex( pathIn = "", dataFile = Files[ "block_3sp.csv" ],
lib = "1 99", pred = "100 198", E = 3, Tp = 1,
knn = 0, tau = -1, exclusionRadius = 0,
columns = "x_t y_t z_t", target = "x_t",
embedded = False, showPlot = True,
validLib = [], generateSteps = 0, parameterList = False )
M = EDM.Multiview( pathIn = "", dataFile = Files[ "block_3sp.csv" ],
lib = "1 100", pred = "101 198",
D = 0, E = 3, Tp = 1, knn = 0, tau = -1,
multiview = 0, exclusionRadius = 0,
columns = "x_t y_t z_t", target = "x_t",
trainLib = True, excludeTarget = False,
numThreads = 4, showPlot = True )
# S-map circle : specify multivariable columns embedded = True
S = EDM.SMap( pathIn = "", dataFile = Files[ "circle.csv" ],
lib = "1 100", pred = "101 198",
E = 2, Tp = 1, knn = 0, tau = -1,
theta = 4, exclusionRadius = 0,
columns = "x y", target = "x",
solver = None, embedded = True,
validLib = [], generateSteps = 0, parameterList = False )
CM = EDM.CCM( pathIn = "", dataFile = Files[ "sardine_anchovy_sst.csv" ],
E = 3, Tp = 0, knn = 0, tau = -1, exclusionRadius = 0,
columns = "anchovy", target = "np_sst",
libSizes = "10 70 10", sample = 100, random = True,
replacement = False, seed = 0, includeData = False,
parameterList = False, verbose = False, showPlot = True )
