idfk anymore statistics or something

src/experiments/anova.py

@@ -0,0 +1,68 @@
+from scipy.stats import f_oneway
+import csv
+import numpy as np
+from enum import Enum
+import pandas as pd 
+import statsmodels.api as sm 
+from statsmodels.formula.api import ols
+from IPython.display import display
+
+class Tree(Enum):
+    ACCASIA = 0
+    BERK = 1
+    EIK = 2
+    ELS = 3
+    ESDOORN = 4
+    ES = 5
+    LINDE = 6
+    PLATAAN = 7
+
+PATH = "C:\\Users\\Tom\\Desktop\\Files\\Repositories\\EV5_Beeldherk_Bomen\\dataset\\csv\\complete.csv"
+
+data = []
+with open(PATH) as file:
+    reader = csv.reader(file, delimiter=',')
+    lst = list(reader)
+    for row in lst:
+        data.append(row)
+
+header = data.pop(0)
+header = [x.strip() for x in header]
+
+# Create a dataframe 
+df = pd.DataFrame(data, columns=header)
+df.astype({col: float for col in df.columns[1:]})
+
+df['Tree'] = [Tree[x.upper()].value for x in df['Tree']]
+
+for idx, name in enumerate(header[1:]):
+    df[name] = df[name].astype(float)
+    # df[name] = (df[name] - df[name].mean()) / df[name].std()
+
+# df.info()
+
+
+
+result = {}
+for name in header:
+    prep = [[] for x in range(8)]
+    for idx, tree in enumerate(df['Tree']):
+        prep[tree].append(df[name][idx])
+
+    F, p = f_oneway(
+        prep[0], prep[1],prep[2], prep[3], 
+        prep[4], prep[5], prep[6], prep[7])
+    result[name]= p
+
+result = sorted(result.items(), key=lambda x:x[1])
+
+print("p score")
+for res in result[1:]:
+    print(res[0], res[1])
+
+# df.boxplot(header[29], by="Tree", figsize=(17, 24))
+
+# for idx, name in enumerate(header[1:]):
+#     F, p = f_oneway(df['Tree'], df[name])
+#     if p < 0.05 and p != 0:
+#         print(idx+1, name, '\t\t', p)