refractoring: added currying
This commit is contained in:
Sandro Zimmermann
parent
2becf5823b
commit
b08b3aba26
180
main.py
180
main.py
@ -1,84 +1,11 @@
|
||||
import sys
|
||||
import pandas as pd
|
||||
import numpy as np
|
||||
import seaborn as sns
|
||||
import matplotlib.pyplot as plt
|
||||
from sklearn.metrics import f1_score
|
||||
from sklearn.model_selection import train_test_split
|
||||
from sklearn.ensemble import RandomForestClassifier
|
||||
from sklearn.neighbors import KNeighborsClassifier
|
||||
from sklearn.tree import DecisionTreeClassifier
|
||||
from sklearn.preprocessing import LabelEncoder
|
||||
from py.arguments import Arguments
|
||||
from py.modell import *
|
||||
|
||||
if not sys.argv[1:]:
|
||||
print("Usage: python3 main.py <path to csv>")
|
||||
sys.exit(1)
|
||||
|
||||
FEATURES = ["points", "x", "y"]
|
||||
|
||||
# create dataframe from csv and drop any row with null values
|
||||
def load_dataframe(file_path):
|
||||
try:
|
||||
colum_list = FEATURES
|
||||
df = pd.read_csv(file_path, usecols = colum_list).dropna()
|
||||
return df
|
||||
except FileNotFoundError as error:
|
||||
print(error)
|
||||
quit()
|
||||
|
||||
def calc_f1_macro(y_true, y_pred):
|
||||
f1_scores = []
|
||||
for column in y_true:
|
||||
score = calc_f1_score(y_true[column].values, y_pred[column])
|
||||
f1_scores.append(score)
|
||||
return np.mean(f1_scores)
|
||||
|
||||
def calc_f1_score(y_true, y_pred):
|
||||
tp = np.sum(np.multiply([i==True for i in y_pred], y_true))
|
||||
tn = np.sum(np.multiply([i==False for i in y_pred], [not(j) for j in y_true]))
|
||||
fp = np.sum(np.multiply([i==True for i in y_pred], [not(j) for j in y_true]))
|
||||
fn = np.sum(np.multiply([i==False for i in y_pred], y_true))
|
||||
|
||||
precision = calc_precision(tp, fp)
|
||||
recall = calc_recall(tp, fn)
|
||||
|
||||
'''
|
||||
if tp != 0 and fp != 0:
|
||||
precision = calc_precision(tp, fp)
|
||||
else:
|
||||
precision = 0
|
||||
|
||||
if tp != 0 and fn != 0:
|
||||
recall = calc_recall(tp, fn)
|
||||
else:
|
||||
recall = 0
|
||||
'''
|
||||
|
||||
if precision != 0 and recall != 0:
|
||||
f1 = (2 * precision * recall) / (precision + recall)
|
||||
else:
|
||||
f1 = 0
|
||||
return f1
|
||||
|
||||
def calc_precision(tp, fp):
|
||||
return tp / (tp + fp)
|
||||
|
||||
def calc_recall(tp, fn):
|
||||
return tp / (tp + fn)
|
||||
|
||||
def get_score_from_cli(mode):
|
||||
try:
|
||||
x = float(input("x: "))
|
||||
y = float(input("y: "))
|
||||
if mode == "v":
|
||||
abs_v = np.sqrt(x**2 + y**2)
|
||||
return np.array([abs_v]).reshape(1, -1)
|
||||
return np.array([x, y]).reshape(1, -1)
|
||||
except ValueError:
|
||||
print("Invalid input. Please enter numeric values.")
|
||||
return None
|
||||
|
||||
def main():
|
||||
|
||||
repeat = True
|
||||
@ -87,103 +14,50 @@ def main():
|
||||
args.set_information(False)
|
||||
args.set_graph(False)
|
||||
|
||||
settings = {
|
||||
"repeat": True,
|
||||
"file": args.get_file_path(),
|
||||
"mode": args.get_mode(),
|
||||
"information": args.get_information(),
|
||||
"graph": args.get_graph()
|
||||
}
|
||||
|
||||
while repeat:
|
||||
print("Currently selected setting:")
|
||||
print(f"File: {settings["file"]}")
|
||||
print(f"Mode: {settings["mode"]}")
|
||||
print(f"Display information: {settings["information"]}")
|
||||
print(f"Display graphs: {settings["graph"]}")
|
||||
print(f"File: {args.get_file_path()}")
|
||||
print(f"Mode: {args.get_mode()}")
|
||||
print(f"Display information: {args.get_information()}")
|
||||
print(f"Display graphs: {args.get_graph()}")
|
||||
|
||||
prompt = input("Change settings [y / exit]: ")
|
||||
prompt = input("Change settings [y / exit / blank]: ")
|
||||
|
||||
if prompt == "y":
|
||||
try:
|
||||
args.set_file_path(input("Change file <path to file>: "))
|
||||
args.set_mode(input("Change mode [v, a, c]: "))
|
||||
args.set_information(bool(input("Display information [True / False]: ")))
|
||||
args.set_graph(bool(input("Display graphs [True / False]: ")))
|
||||
args.set_information(eval(input("Display information [True / False]: ")))
|
||||
args.set_graph(eval(input("Display graphs [True / False]: ")))
|
||||
except ValueError as error:
|
||||
print(f"Value {error}")
|
||||
elif prompt == "exit":
|
||||
quit()
|
||||
|
||||
# load dataframe with argument [1]
|
||||
df = load_dataframe(args.get_file_path())
|
||||
|
||||
# print dataframe information if argument [3] is true
|
||||
if args.get_information():
|
||||
print(df.describe())
|
||||
print(df.head())
|
||||
print(df.head().info())
|
||||
|
||||
# display graphs if argument [4] is true
|
||||
if args.get_graph():
|
||||
|
||||
sns.countplot(x = df["points"])
|
||||
plt.show()
|
||||
|
||||
sns.heatmap(df.corr(), annot=True, cmap='coolwarm')
|
||||
plt.show()
|
||||
|
||||
sns.scatterplot(x=df['x'], y=df['y'], hue=df['points'])
|
||||
plt.show()
|
||||
|
||||
# use verctor length of (x,y) as feature
|
||||
if args.get_mode() == "v":
|
||||
df["radius"] = np.sqrt(df["x"]**2 + df["y"]**2)
|
||||
X = df[["radius"]]
|
||||
load_dataframe = make_dataframe(lambda df: df)
|
||||
features = make_features(radius)
|
||||
score = make_score_function(lambda x, y: [np.sqrt(x**2 + y**2)])
|
||||
# use absoult values of (x,y) as feature
|
||||
elif args.get_mode() == "a":
|
||||
df_abs = df.copy().abs()
|
||||
features = ["x", "y"]
|
||||
X = df[features]
|
||||
load_dataframe = make_dataframe(lambda df: df.abs())
|
||||
features = make_features(xy)
|
||||
score = make_score_function(lambda x, y: [x, y])
|
||||
# use unaltered values of (x,y) as feature
|
||||
elif args.get_mode() == "c":
|
||||
features = ["x", "y"]
|
||||
X = df[features]
|
||||
load_dataframe = make_dataframe(lambda df: df)
|
||||
features = make_features(xy)
|
||||
score = make_score_function(lambda x, y: [x, y])
|
||||
# default use vector length
|
||||
else:
|
||||
load_dataframe = make_dataframe(lambda df: df)
|
||||
features = make_features(radius)
|
||||
score = make_score_function(lambda x, y: [np.sqrt(x**2 + y**2)])
|
||||
|
||||
y = pd.get_dummies(df['points'])
|
||||
|
||||
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=0)
|
||||
|
||||
random_forest = RandomForestClassifier(n_estimators=700, random_state=0)
|
||||
decision_tree = DecisionTreeClassifier(random_state=0)
|
||||
k_neighbors = KNeighborsClassifier(n_neighbors=5)
|
||||
|
||||
models = {
|
||||
"Random Forest Classifier": random_forest,
|
||||
"Decision Tree Classifier": decision_tree,
|
||||
"K-Neighbors": k_neighbors
|
||||
}
|
||||
|
||||
for name, model in models.items():
|
||||
model.fit(X_train.values, y_train.values)
|
||||
|
||||
for name, model in models.items():
|
||||
pred = model.predict(X_test.values)
|
||||
|
||||
my_f1_macro_score = calc_f1_macro(y_test, pd.DataFrame(pred))
|
||||
print(f'My F1 score of {name} is {my_f1_macro_score}\n')
|
||||
|
||||
f1_sklearn = f1_score(y_test.values, pred, average='macro')
|
||||
print(f'Sklearn F1 score of {name} is {f1_sklearn}\n')
|
||||
|
||||
score = get_score_from_cli(args.get_mode)
|
||||
|
||||
label_encoder = LabelEncoder()
|
||||
df["points"] = label_encoder.fit_transform(df["points"])
|
||||
|
||||
for name, model in models.items():
|
||||
pred = model.predict(score)
|
||||
points_number = pd.DataFrame(pred).idxmax(axis=1)
|
||||
points = label_encoder.inverse_transform(points_number)[0]
|
||||
print(f"{name}: {points} Punkte")
|
||||
print("\n")
|
||||
|
||||
apply_model(load_dataframe(args.get_file_path()), features, score, args.get_information(), args.get_graph())
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
@ -1,64 +1,15 @@
|
||||
import pandas as pd
|
||||
import numpy as np
|
||||
import seaborn as sns
|
||||
import matplotlib.pyplot as plt
|
||||
from sklearn.metrics import f1_score
|
||||
from sklearn.model_selection import train_test_split
|
||||
from sklearn.ensemble import RandomForestClassifier
|
||||
from sklearn.neighbors import KNeighborsClassifier
|
||||
from sklearn.tree import DecisionTreeClassifier
|
||||
from sklearn.preprocessing import LabelEncoder
|
||||
|
||||
FEATURES = ["points", "x", "y"]
|
||||
from py.my_functions import *
|
||||
|
||||
# create dataframe from csv and drop any row with null values
|
||||
def load_dataframe():
|
||||
def load_dataframe(file_path):
|
||||
try:
|
||||
colum_list = FEATURES
|
||||
df = pd.read_csv("data/synthetic_data.csv", usecols = colum_list).dropna()
|
||||
df = pd.read_csv(file_path, usecols = colum_list).dropna()
|
||||
return df.abs()
|
||||
except FileNotFoundError as error:
|
||||
print(error)
|
||||
quit()
|
||||
|
||||
def calc_f1_macro(y_true, y_pred):
|
||||
f1_scores = []
|
||||
for column in y_true:
|
||||
score = calc_f1_score(y_true[column].values, y_pred[column])
|
||||
f1_scores.append(score)
|
||||
return np.mean(f1_scores)
|
||||
|
||||
def calc_f1_score(y_true, y_pred):
|
||||
tp = np.sum(np.multiply([i==True for i in y_pred], y_true))
|
||||
tn = np.sum(np.multiply([i==False for i in y_pred], [not(j) for j in y_true]))
|
||||
fp = np.sum(np.multiply([i==True for i in y_pred], [not(j) for j in y_true]))
|
||||
fn = np.sum(np.multiply([i==False for i in y_pred], y_true))
|
||||
|
||||
precision = calc_precision(tp, fp)
|
||||
recall = calc_recall(tp, fn)
|
||||
|
||||
'''if tp != 0 and fp != 0:
|
||||
precision = calc_precision(tp, fp)
|
||||
else:
|
||||
precision = 0
|
||||
|
||||
if tp != 0 and fn != 0:
|
||||
recall = calc_recall(tp, fn)
|
||||
else:
|
||||
recall = 0'''
|
||||
|
||||
if precision != 0 and recall != 0:
|
||||
f1 = (2 * precision * recall) / (precision + recall)
|
||||
else:
|
||||
f1 = 0
|
||||
return f1
|
||||
|
||||
def calc_precision(tp, fp):
|
||||
return tp / (tp + fp)
|
||||
|
||||
def calc_recall(tp, fn):
|
||||
return tp / (tp + fn)
|
||||
|
||||
def get_score_from_cli():
|
||||
try:
|
||||
x = float(input("x: "))
|
||||
@ -68,12 +19,19 @@ def get_score_from_cli():
|
||||
print("Invalid input. Please enter numeric values.")
|
||||
return None
|
||||
|
||||
def main():
|
||||
df = load_dataframe()
|
||||
def absolut(file_path, inf, graph):
|
||||
|
||||
# load dataframe with argument [1]
|
||||
df = load_dataframe(file_path)
|
||||
|
||||
# print dataframe information if argument [3] is true
|
||||
if inf:
|
||||
print(df.describe())
|
||||
print(df.head())
|
||||
print(df.head().info())
|
||||
|
||||
# display graphs if argument [4] is true
|
||||
if graph:
|
||||
sns.countplot(x = df["points"])
|
||||
plt.show()
|
||||
|
||||
@ -107,10 +65,10 @@ def main():
|
||||
pred = model.predict(X_test.values)
|
||||
|
||||
my_f1_macro_score = calc_f1_macro(y_test, pd.DataFrame(pred))
|
||||
print(f'My F1 score of {name} is {my_f1_macro_score}\n')
|
||||
print(f'My F1 score of {name} is {my_f1_macro_score}')
|
||||
|
||||
f1_sklearn = f1_score(y_test.values, pred, average='macro')
|
||||
print(f'Sklearn F1 score of {name} is {f1_sklearn}\n')
|
||||
print(f'Sklearn F1 score of {name} is {f1_sklearn}')
|
||||
|
||||
score = get_score_from_cli()
|
||||
|
||||
@ -122,7 +80,3 @@ def main():
|
||||
points_number = pd.DataFrame(pred).idxmax(axis=1)
|
||||
points = label_encoder.inverse_transform(points_number)[0]
|
||||
print(f"{name}: {points} Punkte")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
@ -31,25 +31,16 @@ class Arguments:
|
||||
return self.mode.value
|
||||
|
||||
def set_mode(self, value):
|
||||
try:
|
||||
self.mode = Mode(value)
|
||||
except ValueError:
|
||||
raise ValueError(f"Invalid mode '{value}'. Allowed values: {[m.value for m in Mode]}")
|
||||
|
||||
def get_information(self):
|
||||
return self.information.value
|
||||
|
||||
def set_information(self, value):
|
||||
try:
|
||||
self.information = Information(value)
|
||||
except ValueError:
|
||||
raise ValueError(f"Invalid information '{value}'. Allowed values: {[m.value for m in Information]}")
|
||||
|
||||
def get_graph(self):
|
||||
return self.graph.value
|
||||
|
||||
def set_graph(self, value):
|
||||
try:
|
||||
self.graph = Graph(value)
|
||||
except ValueError:
|
||||
raise ValueError(f"Invalid graph '{value}'. Allowed values: {[m.value for m in Graph]}")
|
||||
|
||||
@ -0,0 +1,82 @@
|
||||
from py.my_functions import *
|
||||
|
||||
# create dataframe from csv and drop any row with null values
|
||||
def load_dataframe(file_path):
|
||||
try:
|
||||
colum_list = FEATURES
|
||||
df = pd.read_csv(file_path, usecols = colum_list).dropna()
|
||||
return df
|
||||
except FileNotFoundError as error:
|
||||
print(error)
|
||||
quit()
|
||||
|
||||
def get_score_from_cli():
|
||||
try:
|
||||
x = float(input("x: "))
|
||||
y = float(input("y: "))
|
||||
return np.array([x, y]).reshape(1, -1)
|
||||
except ValueError:
|
||||
print("Invalid input. Please enter numeric values.")
|
||||
return None
|
||||
|
||||
def cartesian(file_path, inf, graph):
|
||||
|
||||
# load dataframe with argument [1]
|
||||
df = load_dataframe(file_path)
|
||||
|
||||
# print dataframe information if argument [3] is true
|
||||
if inf:
|
||||
print(df.describe())
|
||||
print(df.head())
|
||||
print(df.head().info())
|
||||
|
||||
# display graphs if argument [4] is true
|
||||
if graph:
|
||||
sns.countplot(x = df["points"])
|
||||
plt.show()
|
||||
|
||||
sns.heatmap(df.corr(), annot=True, cmap='coolwarm')
|
||||
plt.show()
|
||||
|
||||
sns.scatterplot(x=df['x'], y=df['y'], hue=df['points'])
|
||||
plt.show()
|
||||
|
||||
features = ["x", "y"]
|
||||
X = df[features]
|
||||
|
||||
y = pd.get_dummies(df['points'])
|
||||
|
||||
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=0)
|
||||
|
||||
random_forest = RandomForestClassifier(n_estimators=700, random_state=0)
|
||||
decision_tree = DecisionTreeClassifier(random_state=0)
|
||||
k_neighbors = KNeighborsClassifier(n_neighbors=5)
|
||||
|
||||
models = {
|
||||
"Random Forest Classifier": random_forest,
|
||||
"Decision Tree Classifier": decision_tree,
|
||||
"K-Neighbors": k_neighbors
|
||||
}
|
||||
|
||||
for name, model in models.items():
|
||||
model.fit(X_train.values, y_train.values)
|
||||
|
||||
for name, model in models.items():
|
||||
pred = model.predict(X_test.values)
|
||||
|
||||
my_f1_macro_score = calc_f1_macro(y_test, pd.DataFrame(pred))
|
||||
print(f'My F1 score of {name} is {my_f1_macro_score}')
|
||||
|
||||
f1_sklearn = f1_score(y_test.values, pred, average='macro')
|
||||
print(f'Sklearn F1 score of {name} is {f1_sklearn}')
|
||||
|
||||
score = get_score_from_cli()
|
||||
|
||||
label_encoder = LabelEncoder()
|
||||
df["points"] = label_encoder.fit_transform(df["points"])
|
||||
|
||||
for name, model in models.items():
|
||||
pred = model.predict(score)
|
||||
points_number = pd.DataFrame(pred).idxmax(axis=1)
|
||||
points = label_encoder.inverse_transform(points_number)[0]
|
||||
print(f"{name}: {points} Punkte")
|
||||
137
py/modell.py
Normal file
137
py/modell.py
Normal file
@ -0,0 +1,137 @@
|
||||
import pandas as pd
|
||||
import numpy as np
|
||||
import seaborn as sns
|
||||
import matplotlib.pyplot as plt
|
||||
from sklearn.metrics import f1_score
|
||||
from sklearn.model_selection import train_test_split
|
||||
from sklearn.ensemble import RandomForestClassifier
|
||||
from sklearn.neighbors import KNeighborsClassifier
|
||||
from sklearn.tree import DecisionTreeClassifier
|
||||
from sklearn.preprocessing import LabelEncoder
|
||||
|
||||
# low amounts of features will result in many zero devision in tp=0 and fp=0
|
||||
np.seterr(divide='ignore', invalid='ignore')
|
||||
|
||||
FEATURES = ["points", "x", "y"]
|
||||
|
||||
def make_dataframe(transform):
|
||||
def load_dataframe(file_path):
|
||||
try:
|
||||
colum_list = FEATURES
|
||||
df = pd.read_csv(file_path, usecols = colum_list).dropna()
|
||||
return transform(df)
|
||||
except FileNotFoundError as error:
|
||||
print(error)
|
||||
quit()
|
||||
return load_dataframe
|
||||
|
||||
def make_features(selector):
|
||||
def select(df):
|
||||
return df
|
||||
return select(selector)
|
||||
|
||||
def radius(df):
|
||||
df["radius"] = np.sqrt(df["x"]**2 + df["y"]**2)
|
||||
return df[["radius"]]
|
||||
|
||||
def xy(df):
|
||||
features = ["x", "y"]
|
||||
return df[features]
|
||||
|
||||
def apply_model(df, features, score, inf, graph):
|
||||
# print dataframe information
|
||||
if inf:
|
||||
print(df.describe())
|
||||
print(df.head())
|
||||
print(df.head().info())
|
||||
|
||||
# display graphs
|
||||
if graph:
|
||||
sns.countplot(x = df["points"])
|
||||
plt.show()
|
||||
|
||||
sns.heatmap(df.corr(), annot=True, cmap='coolwarm')
|
||||
plt.show()
|
||||
|
||||
sns.scatterplot(x=df['x'], y=df['y'], hue=df['points'])
|
||||
plt.show()
|
||||
|
||||
y = pd.get_dummies(df['points'])
|
||||
X = features(df)
|
||||
|
||||
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=0)
|
||||
|
||||
random_forest = RandomForestClassifier(n_estimators=700, random_state=0)
|
||||
decision_tree = DecisionTreeClassifier(random_state=0)
|
||||
k_neighbors = KNeighborsClassifier(n_neighbors=5)
|
||||
|
||||
models = {
|
||||
"Random Forest Classifier": random_forest,
|
||||
"Decision Tree Classifier": decision_tree,
|
||||
"K-Neighbors": k_neighbors
|
||||
}
|
||||
|
||||
for name, model in models.items():
|
||||
model.fit(X_train.values, y_train.values)
|
||||
|
||||
for name, model in models.items():
|
||||
pred = model.predict(X_test.values)
|
||||
|
||||
my_f1_macro_score = calc_f1_macro(y_test, pd.DataFrame(pred))
|
||||
print(f'My F1 score of {name} is {my_f1_macro_score}')
|
||||
|
||||
f1_sklearn = f1_score(y_test.values, pred, average='macro')
|
||||
print(f'Sklearn F1 score of {name} is {f1_sklearn}')
|
||||
|
||||
score = score()
|
||||
|
||||
label_encoder = LabelEncoder()
|
||||
df["points"] = label_encoder.fit_transform(df["points"])
|
||||
|
||||
for name, model in models.items():
|
||||
pred = model.predict(score)
|
||||
points_number = pd.DataFrame(pred).idxmax(axis=1)
|
||||
points = label_encoder.inverse_transform(points_number)[0]
|
||||
print(f"{name}: {points} Punkte")
|
||||
|
||||
input("\nPress any key to continue...\n")
|
||||
|
||||
# calc f1 macro
|
||||
def calc_f1_macro(y_true, y_pred):
|
||||
f1_scores = []
|
||||
for column in y_true:
|
||||
score = calc_f1_score(y_true[column].values, y_pred[column])
|
||||
f1_scores.append(score)
|
||||
return np.mean(f1_scores)
|
||||
|
||||
def calc_f1_score(y_true, y_pred):
|
||||
tp = np.sum(np.multiply([i==True for i in y_pred], y_true))
|
||||
tn = np.sum(np.multiply([i==False for i in y_pred], [not(j) for j in y_true]))
|
||||
fp = np.sum(np.multiply([i==True for i in y_pred], [not(j) for j in y_true]))
|
||||
fn = np.sum(np.multiply([i==False for i in y_pred], y_true))
|
||||
|
||||
precision = calc_precision(tp, fp)
|
||||
recall = calc_recall(tp, fn)
|
||||
|
||||
if precision != 0 and recall != 0:
|
||||
f1 = (2 * precision * recall) / (precision + recall)
|
||||
else:
|
||||
f1 = 0
|
||||
return f1
|
||||
|
||||
def calc_precision(tp, fp):
|
||||
return tp / (tp + fp)
|
||||
|
||||
def calc_recall(tp, fn):
|
||||
return tp / (tp + fn)
|
||||
|
||||
def make_score_function(transform):
|
||||
def get_score_from_cli():
|
||||
try:
|
||||
x = float(input("x: "))
|
||||
y = float(input("y: "))
|
||||
return np.array([transform(x, y)]).reshape(1, -1)
|
||||
except ValueError:
|
||||
print("Invalid input. Please enter numeric values.")
|
||||
return None
|
||||
return get_score_from_cli
|
||||
88
py/my_functions.py
Normal file
88
py/my_functions.py
Normal file
@ -0,0 +1,88 @@
|
||||
import pandas as pd
|
||||
import numpy as np
|
||||
import seaborn as sns
|
||||
import matplotlib.pyplot as plt
|
||||
from sklearn.metrics import f1_score
|
||||
from sklearn.model_selection import train_test_split
|
||||
from sklearn.ensemble import RandomForestClassifier
|
||||
from sklearn.neighbors import KNeighborsClassifier
|
||||
from sklearn.tree import DecisionTreeClassifier
|
||||
from sklearn.preprocessing import LabelEncoder
|
||||
|
||||
np.seterr(divide='ignore', invalid='ignore')
|
||||
|
||||
FEATURES = ["points", "x", "y"]
|
||||
|
||||
def make_dataframe(transform):
|
||||
def load_dataframe(file_path):
|
||||
try:
|
||||
colum_list = FEATURES
|
||||
df = pd.read_csv(file_path, usecols = colum_list).dropna()
|
||||
return transform(df)
|
||||
except FileNotFoundError as error:
|
||||
print(error)
|
||||
quit()
|
||||
|
||||
def calc_f1_macro(y_true, y_pred):
|
||||
f1_scores = []
|
||||
for column in y_true:
|
||||
score = calc_f1_score(y_true[column].values, y_pred[column])
|
||||
f1_scores.append(score)
|
||||
return np.mean(f1_scores)
|
||||
|
||||
def calc_f1_score(y_true, y_pred):
|
||||
tp = np.sum(np.multiply([i==True for i in y_pred], y_true))
|
||||
tn = np.sum(np.multiply([i==False for i in y_pred], [not(j) for j in y_true]))
|
||||
fp = np.sum(np.multiply([i==True for i in y_pred], [not(j) for j in y_true]))
|
||||
fn = np.sum(np.multiply([i==False for i in y_pred], y_true))
|
||||
|
||||
precision = calc_precision(tp, fp)
|
||||
recall = calc_recall(tp, fn)
|
||||
|
||||
if precision != 0 and recall != 0:
|
||||
f1 = (2 * precision * recall) / (precision + recall)
|
||||
else:
|
||||
f1 = 0
|
||||
return f1
|
||||
|
||||
def calc_precision(tp, fp):
|
||||
return tp / (tp + fp)
|
||||
|
||||
def calc_recall(tp, fn):
|
||||
return tp / (tp + fn)
|
||||
|
||||
def apply_model(X, y):
|
||||
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=0)
|
||||
|
||||
random_forest = RandomForestClassifier(n_estimators=700, random_state=0)
|
||||
decision_tree = DecisionTreeClassifier(random_state=0)
|
||||
k_neighbors = KNeighborsClassifier(n_neighbors=5)
|
||||
|
||||
models = {
|
||||
"Random Forest Classifier": random_forest,
|
||||
"Decision Tree Classifier": decision_tree,
|
||||
"K-Neighbors": k_neighbors
|
||||
}
|
||||
|
||||
for name, model in models.items():
|
||||
model.fit(X_train.values, y_train.values)
|
||||
|
||||
for name, model in models.items():
|
||||
pred = model.predict(X_test.values)
|
||||
|
||||
my_f1_macro_score = calc_f1_macro(y_test, pd.DataFrame(pred))
|
||||
print(f'My F1 score of {name} is {my_f1_macro_score}')
|
||||
|
||||
f1_sklearn = f1_score(y_test.values, pred, average='macro')
|
||||
print(f'Sklearn F1 score of {name} is {f1_sklearn}')
|
||||
|
||||
def make_score_function(transform):
|
||||
def get_score_from_cli():
|
||||
try:
|
||||
x = float(input("x: "))
|
||||
y = float(input("y: "))
|
||||
return np.array([transform(x, y)]).reshape(1, -1)
|
||||
except ValueError:
|
||||
print("Invalid input. Please enter numeric values.")
|
||||
return None
|
||||
return get_score_from_cli
|
||||
81
py/vector.py
81
py/vector.py
@ -1,64 +1,15 @@
|
||||
import pandas as pd
|
||||
import numpy as np
|
||||
import seaborn as sns
|
||||
import matplotlib.pyplot as plt
|
||||
from sklearn.metrics import f1_score
|
||||
from sklearn.model_selection import train_test_split
|
||||
from sklearn.ensemble import RandomForestClassifier
|
||||
from sklearn.neighbors import KNeighborsClassifier
|
||||
from sklearn.tree import DecisionTreeClassifier
|
||||
from sklearn.preprocessing import LabelEncoder
|
||||
|
||||
FEATURES = ["points", "x", "y"]
|
||||
from my_functions import *
|
||||
|
||||
# create dataframe from csv and drop any row with null values
|
||||
def load_dataframe():
|
||||
def load_dataframe(file_path):
|
||||
try:
|
||||
colum_list = FEATURES
|
||||
df = pd.read_csv("data/synthetic_data.csv", usecols = colum_list).dropna()
|
||||
df = pd.read_csv(file_path, usecols = colum_list).dropna()
|
||||
return df
|
||||
except FileNotFoundError as error:
|
||||
print(error)
|
||||
quit()
|
||||
|
||||
def calc_f1_macro(y_true, y_pred):
|
||||
f1_scores = []
|
||||
for column in y_true:
|
||||
score = calc_f1_score(y_true[column].values, y_pred[column])
|
||||
f1_scores.append(score)
|
||||
return np.mean(f1_scores)
|
||||
|
||||
def calc_f1_score(y_true, y_pred):
|
||||
tp = np.sum(np.multiply([i==True for i in y_pred], y_true))
|
||||
tn = np.sum(np.multiply([i==False for i in y_pred], [not(j) for j in y_true]))
|
||||
fp = np.sum(np.multiply([i==True for i in y_pred], [not(j) for j in y_true]))
|
||||
fn = np.sum(np.multiply([i==False for i in y_pred], y_true))
|
||||
|
||||
precision = calc_precision(tp, fp)
|
||||
recall = calc_recall(tp, fn)
|
||||
|
||||
'''if tp != 0 and fp != 0:
|
||||
precision = calc_precision(tp, fp)
|
||||
else:
|
||||
precision = 0
|
||||
|
||||
if tp != 0 and fn != 0:
|
||||
recall = calc_recall(tp, fn)
|
||||
else:
|
||||
recall = 0'''
|
||||
|
||||
if precision != 0 and recall != 0:
|
||||
f1 = (2 * precision * recall) / (precision + recall)
|
||||
else:
|
||||
f1 = 0
|
||||
return f1
|
||||
|
||||
def calc_precision(tp, fp):
|
||||
return tp / (tp + fp)
|
||||
|
||||
def calc_recall(tp, fn):
|
||||
return tp / (tp + fn)
|
||||
|
||||
def get_score_from_cli():
|
||||
try:
|
||||
x = float(input("x: "))
|
||||
@ -69,10 +20,19 @@ def get_score_from_cli():
|
||||
print("Invalid input. Please enter numeric values.")
|
||||
return None
|
||||
|
||||
def main():
|
||||
df = load_dataframe()
|
||||
print(df.head())
|
||||
def vector(file_path, inf, graph):
|
||||
|
||||
# load dataframe
|
||||
df = load_dataframe(file_path)
|
||||
|
||||
# print dataframe information
|
||||
if inf:
|
||||
print(df.describe())
|
||||
print(df.head())
|
||||
print(df.head().info())
|
||||
|
||||
# display graphs
|
||||
if graph:
|
||||
sns.countplot(x = df["points"])
|
||||
plt.show()
|
||||
|
||||
@ -89,6 +49,9 @@ def main():
|
||||
|
||||
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=0)
|
||||
|
||||
print(X_train)
|
||||
quit()
|
||||
|
||||
random_forest = RandomForestClassifier(n_estimators=700, random_state=0)
|
||||
decision_tree = DecisionTreeClassifier(random_state=0)
|
||||
k_neighbors = KNeighborsClassifier(n_neighbors=5)
|
||||
@ -106,10 +69,10 @@ def main():
|
||||
pred = model.predict(X_test.values)
|
||||
|
||||
my_f1_macro_score = calc_f1_macro(y_test, pd.DataFrame(pred))
|
||||
print(f'My F1 score of {name} is {my_f1_macro_score}\n')
|
||||
print(f'My F1 score of {name} is {my_f1_macro_score}')
|
||||
|
||||
f1_sklearn = f1_score(y_test.values, pred, average='macro')
|
||||
print(f'Sklearn F1 score of {name} is {f1_sklearn}\n')
|
||||
print(f'Sklearn F1 score of {name} is {f1_sklearn}')
|
||||
|
||||
score = get_score_from_cli()
|
||||
|
||||
@ -122,6 +85,4 @@ def main():
|
||||
points = label_encoder.inverse_transform(points_number)[0]
|
||||
print(f"{name}: {points} Punkte")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
vector("data/shots.csv", False, False)
|
||||
Loading…
x
Reference in New Issue
Block a user