report = ProfileReport(arbis_matched,
title='ArbIS Matched Dataset Report')
report.to_file(work_path + file_prefix + '_report.html')
###################
### Correlation ###
###################
# define column types
nominal_columns = ['Strasse', 'Month']
dichotomous_columns = ['Richtung']
ordinal_columns = ['AnzGesperrtFs', 'Einzug']
# Encode non numerical columns
arbis_encoded, arbis_encoded_dict = numerical_encoding(
arbis_matched, ["Strasse", 'Month'],
drop_single_label=False,
drop_fact_dict=False)
arbis_encoded.to_csv(csv_path + 'encoded.csv', index=False, sep=';')
with open(csv_path + 'encoded_dict.csv', 'w') as tf:
for key in arbis_encoded_dict.keys():
tf.write("%s, %s\n" % (key, arbis_encoded_dict[key]))
# Calculate with Cramers 's V
results = None # To make sure that no old data is reused
results = compute_correlations(arbis_encoded,
columns_nominal=nominal_columns,
columns_dichotomous=dichotomous_columns,
columns_ordinal=ordinal_columns,
bias_correction=False)
###################
### Encoding ###
###################
# define column types
nominal_columns = [
"Str", "Kat", "Typ", "UArt1", "UArt2", "AUrs1", "AUrs2", "AufHi",
"Char1", "Char2", "Bes1", "Bes2", "Lich1", "Lich2", "Zust1", "Zust2",
"WoTag", 'Month'
]
dichotomous_columns = ["Alkoh"]
ordinal_columns = ["Betei", "Fstf", "FeiTag"]
# Encode non numerical columns
baysis_encoded, baysis_encoded_dict = numerical_encoding(
baysis_selected, ["Strasse", "Fstf", 'Month'],
drop_single_label=False,
drop_fact_dict=False)
baysis_encoded.to_csv(csv_path + 'encoded.csv', index=False, sep=';')
with open(csv_path + 'encoded_dict.csv', 'w') as tf:
for key in baysis_encoded_dict.keys():
tf.write("%s, %s\n" % (key, baysis_encoded_dict[key]))
###################
### Correlation ###
###################
baysis_encoded = baysis_encoded.rename(
columns={
"TempMax": "TMax",
"TempAvg": "TAvg",
###################
### Correlation ###
###################
# define column types
nominal_columns = [
'Str', 'Kat', 'Typ', 'UArt1', 'UArt2', 'AUrs1', 'AUrs2', 'AufHi',
'Char1', 'Char2', 'Lich1', 'Lich2', 'Zust1', 'Zust2', 'WoTag', 'Month'
]
dichotomous_columns = ['Alkoh']
ordinal_columns = ['Betei', 'Fstf', 'FeiTag']
# Encode non numerical columns
baysis_encoded, baysis_encoded_dict = numerical_encoding(
baysis_matched, ['Strasse', 'Fstf', 'WoTag', 'Month'],
drop_single_label=False,
drop_fact_dict=False)
baysis_encoded.to_csv(csv_path + 'encoded.csv', index=False, sep=';')
with open(csv_path + 'encoded_dict.csv', 'w') as tf:
for key in baysis_encoded_dict.keys():
tf.write("%s, %s\n" % (key, baysis_encoded_dict[key]))
baysis_encoded = baysis_encoded.rename(
columns={
"TempMax": "TMax",
"TempAvg": "TAvg",
"SpatMax": "SMax",
"SpatAvg": "SAvg",
"Coverage": "Cov",
"TempDist": "TDist",
"Fstf",
"WoTag",
"FeiTag",
'Month'
]
dichotomous_columns = []
ordinal_columns = []
# Encode non numerical columns
baysis_encoded, baysis_encoded_dict = numerical_encoding(baysis_selected,
["Strasse",
"Fstf",
'Month',
"TempMax",
"TempAvg",
"SpatMax",
"SpatAvg",
"Coverage",
"TLCar",
"TLHGV"
],
drop_single_label=False,
drop_fact_dict=False)
baysis_encoded.to_csv(csv_path + 'encoded.csv', index=False, sep=';')
with open(csv_path + 'encoded_dict.csv', 'w') as tf:
for key in baysis_encoded_dict.keys():
tf.write("%s, %s\n" % (key, baysis_encoded_dict[key]))
###################
### Correlation ###
###################
"Str",
"AGF",
"Einzug",
"Richtung",
"Length",
"Duration",
'Month'
]
dichotomous_columns = []
ordinal_columns = []
# Encode non numerical columns
arbis_encoded, arbis_encoded_dict = numerical_encoding(
arbis_selected, [
"Strasse", 'Month', "Length", "Duration", "TempMax", "TempAvg",
"SpatMax", "SpatAvg", "Coverage", "TLCar", "TLHGV"
],
drop_single_label=False,
drop_fact_dict=False)
arbis_encoded.to_csv(csv_path + 'encoded.csv', index=False, sep=';')
with open(csv_path + 'encoded_dict.csv', 'w') as tf:
for key in arbis_encoded_dict.keys():
tf.write("%s, %s\n" % (key, arbis_encoded_dict[key]))
arbis_encoded = arbis_encoded.rename(
columns={
"TempMax": "TMax",
"TempAvg": "TAvg",
"SpatMax": "SMax",
"SpatAvg": "SAvg",
"AufHi",
"Char1", "Char2",
"Lich1", "Lich2",
"Zust1", "Zust2",
"WoTag",
"FeiTag", 'Month']
dichotomous_columns = ["Alkoh"]
ordinal_columns = ["Betei", "Fstf"]
# Encode non numerical columns
baysis_encoded, baysis_encoded_dict = numerical_encoding(baysis_original,
["Strasse", "Kat", "Typ",
"UArt1", "UArt2",
"AUrs1", "AUrs2",
"AufHi",
"Char1", "Char2",
"Lich1", "Lich2",
"Zust1", "Zust2",
"Fstf"
"WoTag",
"FeiTag", 'Month'], drop_single_label=False,
drop_fact_dict=False)
baysis_encoded.to_csv(csv_path + 'encoded.csv', index=False, sep=';')
with open(csv_path + 'encoded_dict.csv', 'w') as tf:
for key in baysis_encoded_dict.keys():
tf.write("%s, %s\n" % (key, baysis_encoded_dict[key]))
###################
### Correlation ###
###################
report = ProfileReport(arbis_original,
title='ArbIS Original Dataset Report')
report.to_file(work_path + file_prefix + '_report.html')
###################
### Correlation ###
###################
# define column types
nominal_columns = ['Strasse', 'StreckeID', 'Month']
dichotomous_columns = ['Richtung']
ordinal_columns = ['AnzGesperrtFs', 'Einzug']
# Encode non numerical columns
arbis_encoded, arbis_encoded_dict = numerical_encoding(
arbis_original, ['Strasse', 'StreckeID', 'Month'],
drop_single_label=False,
drop_fact_dict=False)
arbis_encoded.to_csv(csv_path + 'encoded.csv', index=False, sep=';')
with open(csv_path + 'encoded_dict.csv', 'w') as tf:
for key in arbis_encoded_dict.keys():
tf.write("%s, %s\n" % (key, arbis_encoded_dict[key]))
# Calculate with Cramers 's V
results = None # To make sure that no old data is reused
results = compute_correlations(arbis_encoded,
columns_nominal=nominal_columns,
columns_dichotomous=dichotomous_columns,
columns_ordinal=ordinal_columns,
bias_correction=False)
