Random Forest, Hyperparameter Tuning, Feature Importance

Random Forest/.vscode/settings.json

+{
+    "python.pythonPath": "/usr/bin/python3.8"
+}
\ No newline at end of file

Random Forest/grid_search.py

+from sklearn.model_selection import train_test_split
+import numpy as np
+import pandas as pd
+from sklearn.ensemble import RandomForestRegressor
+from sklearn.metrics import r2_score,mean_squared_error,mean_absolute_percentage_error
+from sklearn.model_selection import GridSearchCV
+
+
+dataset = np.genfromtxt('data.csv', delimiter=',', skip_header=1)
+x = dataset[:,[0,1,2,4,5,6,7,8,9]]
+y = dataset[:,3]
+x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.25, random_state=42)
+
+
+param_grid = {
+
+    'random_state': [42]
+}
+
+# Use the random grid to search for best hyperparameters
+# First create the base model to tune
+rf = RandomForestRegressor()
+# Random search of parameters, using 3 fold cross validation, 
+# search across 100 different combinations, and use all available cores
+grid_search = GridSearchCV(estimator = rf, param_grid = param_grid, cv = 3, n_jobs = -1, verbose = 2)
+grid_search.fit(x_train, y_train)
+
+print(grid_search.best_params_)
+
+y_predict = grid_search.best_estimator_.predict(x_test)
+accuracy = (1-mean_absolute_percentage_error(y_test, y_predict))*100
+print("Accuracy: " + str(accuracy) + '%')
+
+

Random Forest/param.txt

+{'bootstrap': True, 'max_depth': 100, 'min_samples_leaf': 1, 'min_samples_split': 2, 'n_estimators': 1000}
+Accuracy: 98.859838775608%
\ No newline at end of file

Random Forest/prepare_data.py

+import pandas as pd
+
+df = pd.read_csv("db_dump.csv", delimiter=';', usecols=[
+                 'Streckenvorhersage.Dauer', 'Streckenvorhersage.StartortID', 'Streckenvorhersage.ZielortID', 'Streckenvorhersage.Entfernung', 'Streckenvorhersage.Datum'])
+
+df['Streckenvorhersage.Year'] = [pd.Timestamp(
+    d).year for d in df['Streckenvorhersage.Datum']]
+df['Streckenvorhersage.Month'] = [pd.Timestamp(
+    d).month for d in df['Streckenvorhersage.Datum']]
+df['Streckenvorhersage.Day'] = [pd.Timestamp(
+    d).day for d in df['Streckenvorhersage.Datum']]
+df['Streckenvorhersage.DayOfWeek'] = [pd.Timestamp(
+    d).day_of_week for d in df['Streckenvorhersage.Datum']]
+df['Streckenvorhersage.Hour'] = [pd.Timestamp(
+    d).hour for d in df['Streckenvorhersage.Datum']]
+df['Streckenvorhersage.Minutes'] = [pd.Timestamp(
+    d).minute for d in df['Streckenvorhersage.Datum']]
+
+df.drop(['Streckenvorhersage.Datum'], axis=1, inplace=True)
+
+#df = pd.concat(
+#    [df, pd.get_dummies(df['Streckenvorhersage.DayOfWeek'])], axis=1)
+
+#li = ['Montag', 'Dienstag', 'Mittwoch', 'Donnerstag', 'Freitag', 'Samstag', 'Sonntag']
+#print(pd.get_dummies(li,df['Streckenvorhersage.DayOfWeek'])
+
+#df.rename(index={9: 'Montag', 10: 'Dienstag', 11: 'Mittwoch', 12: 'Donnerstag',
+#                 13: 'Freitag', 14: 'Samstag', 15: 'Sonntag'}, inplace=True)
+#df.drop(['Streckenvorhersage.DayOfWeek'], axis=1, inplace=True)
+
+print(df.columns)
+df.to_csv("data.csv", index=False)

Random Forest/random_forrest.py

+from sklearn.model_selection import train_test_split
+import numpy as np
+import pandas as pd
+from sklearn.ensemble import RandomForestRegressor
+from sklearn.metrics import r2_score,mean_squared_error,mean_absolute_percentage_error
+from pprint import pprint
+import matplotlib.pyplot as plt
+
+dataset = np.genfromtxt('data.csv', delimiter=',', skip_header=1)
+x = dataset[:,[0,1,2,4,5,6,7,8,9]]
+y = dataset[:,3]
+print(x)
+x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.25, random_state=42)
+
+model = RandomForestRegressor(random_state=42)
+model.fit(x_train,y_train)
+pprint(model.get_params())
+
+feature_list = ["StandortID","ZielortID","Entfernung","Jahr","Monat","Tag","Wochentag","Stunde","Minute"]
+importances = np.round(model.feature_importances_ * 100,decimals=2)
+zipped = list(zip(feature_list,importances))
+pprint(zipped)
+
+#plt.style.use('fivethirtyeight')
+x_values = list(range(len(importances)))
+plt.figure(figsize=(10,10))
+plt.bar(x_values, importances, orientation = 'vertical')
+plt.xticks(x_values, feature_list, rotation='vertical')
+plt.ylabel('Importance'); plt.xlabel('Feature'); plt.title('Feature Analyse');
+plt.show()
+
+y_predict = model.predict(x_test)
+accuracy = (1-mean_absolute_percentage_error(y_test, y_predict))*100
+print("Accuracy: " + str(accuracy) + '%')
\ No newline at end of file

Random Forest/random_search.py

+
+from sklearn.model_selection import train_test_split
+import numpy as np
+import pandas as pd
+from sklearn.ensemble import RandomForestRegressor
+from sklearn.metrics import r2_score,mean_squared_error,mean_absolute_percentage_error
+from sklearn.model_selection import GridSearchCV# Create the parameter grid based on the results of random search 
+from sklearn.model_selection import RandomizedSearchCV# Number of trees in random forest
+
+dataset = np.genfromtxt('data.csv', delimiter=',', skip_header=1)
+x = dataset[:,[0,1,2,4,5,6,7,8,9]]
+y = dataset[:,3]
+x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.25, random_state=42)
+
+n_estimators = [int(x) for x in np.linspace(start = 200, stop = 1000, num = 5)]
+# Number of features to consider at every split
+max_features = ['auto', 'sqrt']
+# Maximum number of levels in tree
+max_depth = [int(x) for x in np.linspace(10, 110, num = 11)]
+max_depth.append(None)
+# Minimum number of samples required to split a node
+min_samples_split = [2, 5, 10]
+# Minimum number of samples required at each leaf node
+min_samples_leaf = [1, 2, 4]
+# Method of selecting samples for training each tree
+bootstrap = [True, False]# Create the random grid
+random_grid = {'n_estimators': n_estimators,
+               'max_features': max_features,
+               'max_depth': max_depth,
+               'min_samples_split': min_samples_split,
+               'min_samples_leaf': min_samples_leaf,
+               'bootstrap': bootstrap}
+print(random_grid)
+# Use the random grid to search for best hyperparameters
+# First create the base model to tune
+rf = RandomForestRegressor()
+# Random search of parameters, using 3 fold cross validation, 
+# search across 100 different combinations, and use all available cores
+rf_random = RandomizedSearchCV(estimator = rf, param_distributions = random_grid, n_iter = 100, cv = 3, verbose=2, random_state=42, n_jobs = -1)# Fit the random search model
+rf_random.fit(x_train, y_train)
+rf_random.best_params_
+print(rf_random.best_params_)
\ No newline at end of file