cleaned rpe

.gitignore

@@ -133,3 +133,4 @@ rsconnect/
 
 # datafest
 data_preparation/data/*.csv
+*.big

data_preparation/rpe_processing.py

@@ -1,4 +1,13 @@
 import pandas as pd
+import numpy as np
+
+
+def vectorize_mult(column, dictionary, postfix, df, file=None):
+  newCol = column + postfix
+  df[newCol] = df[column].map(dictionary)
+  if file is not None:
+    df.to_csv('cleaned/{}.csv'.format(file))
+
 
 csv = pd.read_csv("data/rpe.csv")
 training = csv["Training"].unique()
@@ -7,34 +16,45 @@ boms = csv["BestOutOfMyself"].unique()
 print(training)
 print(session)
 print(boms)
+vectorize_mult("Training", {"No": 0, "Yes": 1}, "", csv)
 
+mapping = {"Mobility/Recovery": 1, "Game": 0, "Skills": 0, "Conditioning": 0,
+           "Strength": 0, "Combat": 0, "Speed": 0, np.nan: 0}
+vectorize_mult("SessionType", mapping, "Mobility/Recovery", csv)
+mapping["Mobility/Recovery"] = 0
+mapping["Game"] = 1
+vectorize_mult("SessionType", mapping, "Game", csv)
+mapping["Game"] = 0
+mapping["Skills"] = 1
+vectorize_mult("SessionType", mapping, "Skills", csv)
+mapping["Skills"] = 0
+mapping["Conditioning"] = 1
+vectorize_mult("SessionType", mapping, "Conditioning", csv)
+mapping["Conditioning"] = 0
+mapping["Strength"] = 1
+vectorize_mult("SessionType", mapping, "Strength", csv)
+mapping["Strength"] = 0
+mapping["Combat"] = 1
+vectorize_mult("SessionType", mapping, "Combat", csv)
+mapping["Combat"] = 0
+mapping["Speed"] = 1
+vectorize_mult("SessionType", mapping, "Speed", csv)
+mapping["Speed"] = 0
+mapping[np.nan] = 1
+vectorize_mult("SessionType", mapping, "Unknown", csv)
+mapping[np.nan] = 0
 
-# csv["SessionTypeMobilityRecovery"] = csv["SessionType"]
-# csv.SessionTypeMobilityRecovery[csv.SessionType == 'Mobility/Recovery'] = 1
-# csv.SessionTypeMobilityRecovery[csv.SessionType != 'Mobility/Recovery'] = 0
-#
-# csv["SessionTypeGame"] = csv["SessionTypeMobilityRecovery"]
-# csv.SessionTypeGame[csv.SessionType == 'Game'] = 1
-# csv.SessionTypeGame[csv.SessionType != 'Game'] = 0
-#
-# csv["SessionTypeSkills"] = csv["SessionTypeMobilityRecovery"]
-# csv.SessionTypeSkills[csv.SessionType == 'Skills'] = 1
-# csv.SessionTypeSkills[csv.SessionType != 'Skills'] = 0
-#
-# csv["SessionTypeConditioning"] = csv["SessionTypeMobilityRecovery"]
-# csv.SessionTypeConditioning[csv.SessionType == 'Conditioning'] = 1
-# csv.SessionTypeConditioning[csv.SessionType != 'Conditioning'] = 0
-#
-# csv["SessionTypeStrength"] = csv["SessionTypeMobilityRecovery"]
-# csv.SessionTypeStrength[csv.SessionType == 'Strength'] = 1
-# csv.SessionTypeStrength[csv.SessionType != 'Strength'] = 0
-#
-# csv["SessionTypeCombat"] = csv["SessionTypeMobilityRecovery"]
-# csv.SessionTypeCombat[csv.SessionType == 'Combat'] = 1
-# csv.SessionTypeCombat[csv.SessionType != 'Combat'] = 0
-#
-# csv["SessionTypeSpeed"] = csv["SessionTypeMobilityRecovery"]
-# csv.SessionTypeSpeed[csv.SessionType == 'Speed'] = 1
-# csv.SessionTypeSpeed[csv.SessionType != 'Speed'] = 0
+mapping = {"Not at all": 1, "Absolutely": 0, "Somewhat": 0, np.nan: 0}
+vectorize_mult("BestOutOfMyself", mapping, "NotAtAll", csv)
+mapping["Not at all"] = 0
+mapping["Absolutely"] = 1
+vectorize_mult("BestOutOfMyself", mapping, "Absolutely", csv)
+mapping["Absolutely"] = 0
+mapping["Somewhat"] = 1
+vectorize_mult('BestOutOfMyself', mapping, "Somewhat", csv)
+mapping["Somewhat"] = 0
+mapping[np.nan] = 1
+vectorize_mult('BestOutOfMyself', mapping, "Unknown", csv)
+mapping[np.nan] = 0
 
-print(csv.head())
+csv.to_csv("cleaned/notnormalized_clean_rpe.csv")