from IPython.display import display
from NewsGroup_common import newsgroup_data
from NewsGroup_common import basic_model_test
from NewsGroup_common import saveDataframe
from sklearn.svm import LinearSVC
############### Data Processing #####################
X_train, X_test, y_train, y_test = newsgroup_data.getData()
############### Decision Tree Model #####################
support_vector_classifier = LinearSVC(C=1.0, penalty='l2')
predictions, accuracy, report, confusion_matrix = basic_model_test(
support_vector_classifier, X_train, X_test, y_train, y_test,
"NewsGroup Support Vector Classifier")
################ Cross Validation Hyperparametre Tuning ###############################
print("\nHYPERPARAMETRE TUNING")
hyperparams = {
'C': [0.1, 1, 100, 1000],
'penalty': ['l2'],
'loss': ['hinge', 'squared_hinge'],
'max_iter': [1000, 2000]
}
optimized_model = GridSearchCV(estimator=support_vector_classifier,
param_grid=hyperparams,
n_jobs=1,
from NewsGroup_common import newsgroup_data
from NewsGroup_common import basic_model_test
from NewsGroup_common import saveDataframe
from sklearn.ensemble import RandomForestClassifier
############### Data Processing #####################
X_train, X_test, y_train, y_test = newsgroup_data.getData()
############### Decision Tree Model #####################
random_forest = RandomForestClassifier(n_estimators=100,
criterion='gini',
bootstrap=True)
predictions, accuracy, report, confusion_matrix = basic_model_test(
random_forest, X_train, X_test, y_train, y_test, "NewsGroup Random Forest")
################ Cross Validation Hyperparametre Tuning ###############################
print("\nHYPERPARAMETRE TUNING")
hyperparams = {
'n_estimators': [10, 100, 1000],
'criterion': ['gini', 'entropy']
}
optimized_model = GridSearchCV(estimator=random_forest,
param_grid=hyperparams,
n_jobs=1,
cv=3,
verbose=1,
error_score=1)
from IPython.display import display
from NewsGroup_common import newsgroup_data
from NewsGroup_common import basic_model_test
from NewsGroup_common import saveDataframe
from sklearn import tree
############### Data Processing #####################
X_train, X_test, y_train, y_test = newsgroup_data.getData()
############### Decision Tree Model #####################
decision_tree = tree.DecisionTreeClassifier(max_depth=None,
min_samples_split=2)
predictions, accuracy, report, confusion_matrix = basic_model_test(
decision_tree, X_train, X_test, y_train, y_test, "NewsGroup Decision Tree")
################ Cross Validation Hyperparametre Tuning ###############################
print("\nHYPERPARAMETRE TUNING")
hyperparams = {'min_samples_split': [10, 100, 500], 'max_depth': [2, 20, None]}
optimized_model = GridSearchCV(estimator=decision_tree,
param_grid=hyperparams,
n_jobs=1,
cv=3,
verbose=1,
error_score=1)
optimized_model.fit(X_train, y_train)
from NewsGroup_common import basic_model_test
from NewsGroup_common import saveDataframe
from sklearn.linear_model import LogisticRegression
############### Data Processing #####################
X_train, X_test, y_train, y_test = newsgroup_data.getData()
############### Logistic Regression Model #####################
logistic_regression = LogisticRegression(penalty='l2',
max_iter=300,
C=1,
random_state=42)
predictions, accuracy, report, confusion_matrix = basic_model_test(
logistic_regression, X_train, X_test, y_train, y_test,
"NewsGroup Logistic Regression")
################ Cross Validation Hyperparametre Tuning ###############################
print("\nHYPERPARAMETRE TUNING")
hyperparams = {
'penalty': ['l2'],
'solver': ['sag', 'lbfgs'],
'C': [0.1, 1, 100],
'max_iter': [100, 300]
}
optimized_model = GridSearchCV(estimator=logistic_regression,
param_grid=hyperparams,
n_jobs=1,
import pandas as pd
from IPython.display import display
from NewsGroup_common import newsgroup_data
from NewsGroup_common import basic_model_test
from NewsGroup_common import saveDataframe
from sklearn.ensemble import AdaBoostClassifier
############### Data Processing #####################
X_train, X_test, y_train, y_test = newsgroup_data.getData()
############### AdaBoost Model #####################
adaboost = AdaBoostClassifier(n_estimators=50,learning_rate=1.0)
predictions, accuracy, report, confusion_matrix = basic_model_test(adaboost,X_train, X_test, y_train, y_test,"NewsGroup AdaBoost")
################ Cross Validation Hyperparametre Tuning ###############################
print("\nHYPERPARAMETRE TUNING")
hyperparams = {'n_estimators': [50, 100], 'learning_rate': [0.1, 1.0]}
optimized_model = GridSearchCV(estimator=adaboost, param_grid=hyperparams,
n_jobs=1, cv=3, verbose=1, error_score=1)
optimized_model.fit(X_train, y_train)
print(">>>>> Optimized params")
print(optimized_model.best_params_)
cv_results = optimized_model.cv_results_
from IPython.display import display
from NewsGroup_common import newsgroup_data
from NewsGroup_common import basic_model_test
from NewsGroup_common import saveDataframe
from sklearn.naive_bayes import MultinomialNB
############### Data Processing #####################
X_train, X_test, y_train, y_test = newsgroup_data.getData()
############### Decision Tree Model #####################
naive_bayes = MultinomialNB(alpha=1.0)
predictions, accuracy, report, confusion_matrix = basic_model_test(
naive_bayes, X_train, X_test, y_train, y_test,
"NewsGroup Multinomial Naive Bayes")
################ Cross Validation Hyperparametre Tuning ###############################
print("\nHYPERPARAMETRE TUNING")
hyperparams = {'alpha': (1e0, 1e-2, 1e-4, 1e-10)}
optimized_model = GridSearchCV(estimator=naive_bayes,
param_grid=hyperparams,
n_jobs=1,
cv=3,
verbose=1,
error_score=1)
optimized_model.fit(X_train, y_train)