def main():
"""Tesing the performance of LogisticRegression.
"""
@run_time
def batch():
print("Tesing the performance of LogisticRegression(batch)...")
# Train model
clf = LogisticRegression()
clf.fit(data=data_train, label=label_train, learning_rate=0.1, epochs=1000)
# Model evaluation
model_evaluation(clf, data_test, label_test)
print(clf)
@run_time
def stochastic():
print("Tesing the performance of LogisticRegression(stochastic)...")
# Train model
clf = LogisticRegression()
clf.fit(data=data_train, label=label_train, learning_rate=0.01, epochs=100,
method="stochastic", sample_rate=0.8)
# Model evaluation
model_evaluation(clf, data_test, label_test)
print(clf)
# Load data
data, label = load_breast_cancer()
data = min_max_scale(data)
# Split data randomly, train set rate 70%
data_train, data_test, label_train, label_test = train_test_split(
data, label, random_state=10)
batch()
print()
stochastic()
def main():
@run_time
def batch():
print("Tesing the performance of LinearRegression(batch)...")
# Train model
reg = LinearRegression()
reg.fit(X=X_train, y=y_train, lr=0.1, epochs=1000)
# Model evaluation
get_r2(reg, X_test, y_test)
print(reg)
@run_time
def stochastic():
print("Tesing the performance of LinearRegression(stochastic)...")
# Train model
reg = LinearRegression()
reg.fit(X=X_train,
y=y_train,
lr=0.05,
epochs=50,
method="stochastic",
sample_rate=0.6)
# Model evaluation
get_r2(reg, X_test, y_test)
print(reg)
# Load data
X, y = load_boston_house_prices()
X = min_max_scale(X)
# Split data randomly, train set rate 70%
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=20)
batch()
stochastic()
def main():
print("Tesing the performance of Kmeans...")
# Load data
X, y = load_breast_cancer()
X = min_max_scale(X)
# Train model
est = KMeans()
k = 2
est.fit(X, k)
print()
# Model performance
prob_pos = sum(y) / len(y)
print("Positive probability of X is:%.1f%%.\n" % (prob_pos * 100))
y_hat = est.predict(X)
cluster_pos_tot_cnt = {i: [0, 0] for i in range(k)}
for yi_hat, yi in zip(y_hat, y):
cluster_pos_tot_cnt[yi_hat][0] += yi
cluster_pos_tot_cnt[yi_hat][1] += 1
cluster_prob_pos = {k: v[0] / v[1] for k, v in cluster_pos_tot_cnt.items()}
for i in range(k):
tot_cnt = cluster_pos_tot_cnt[i][1]
prob_pos = cluster_prob_pos[i]
print("Count of elements in cluster %d is:%d." %
(i, tot_cnt))
print("Positive probability of cluster %d is:%.1f%%.\n" %
(i, prob_pos * 100))
def main():
@run_time
def batch():
print("Tesing the performance of LogisticRegression(batch)...")
# Train model
clf = LogisticRegression()
clf.fit(X=X_train, y=y_train, lr=0.1, epochs=1000)
# Model evaluation
model_evaluation(clf, X_test, y_test)
print(clf)
@run_time
def stochastic():
print("Tesing the performance of LogisticRegression(stochastic)...")
# Train model
clf = LogisticRegression()
clf.fit(X=X_train, y=y_train, lr=0.01, epochs=100,
method="stochastic", sample_rate=0.8)
# Model evaluation
model_evaluation(clf, X_test, y_test)
print(clf)
# Load data
X, y = load_breast_cancer()
X = min_max_scale(X)
# Split data randomly, train set rate 70%
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=10)
batch()
print()
stochastic()
def main():
"""Tesing the performance of LinearRegression.
"""
@run_time
def batch():
print("Tesing the performance of LinearRegression(batch)...")
# Train model
reg = LinearRegression()
reg.fit(data=data_train, label=label_train, learning_rate=0.1, epochs=1000)
# Model evaluation
get_r2(reg, data_test, label_test)
print(reg)
@run_time
def stochastic():
print("Tesing the performance of LinearRegression(stochastic)...")
# Train model
reg = LinearRegression()
reg.fit(data=data_train, label=label_train, learning_rate=0.05, epochs=50,
method="stochastic", sample_rate=0.6)
# Model evaluation
get_r2(reg, data_test, label_test)
print(reg)
# Load data
data, label = load_boston_house_prices()
data = min_max_scale(data)
# Split data randomly, train set rate 70%
data_train, data_test, label_train, label_test = train_test_split(
data, label, random_state=20)
batch()
stochastic()
def main():
print("Tesing the performance of KNN regressor...")
# Load data
X, y = load_boston_house_prices()
X = min_max_scale(X)
# Split data randomly, train set rate 70%
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=10)
# Train model
reg = KNeighborsRegressor()
reg.fit(X=X_train, y=y_train, k_neighbors=3)
# Model evaluation
get_r2(reg, X_test, y_test)
def main():
print("Tesing the performance of KNN classifier...")
# Load data
X, y = load_breast_cancer()
X = min_max_scale(X)
# Split data randomly, train set rate 70%
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=20)
# Train model
clf = KNeighborsClassifier()
clf.fit(X_train, y_train, k_neighbors=21)
# Model evaluation
model_evaluation(clf, X_test, y_test)
def main():
print("Tesing the performance of Ridge Regressor(stochastic)...")
# Load data
X, y = load_boston_house_prices()
X = min_max_scale(X)
# Split data randomly, train set rate 70%
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=10)
# Train model
reg = Ridge()
reg.fit(X=X_train, y=y_train, lr=0.001, epochs=1000,
method="stochastic", sample_rate=0.5, alpha=1e-7)
# Model evaluation
get_r2(reg, X_test, y_test)
def main():
"""Tesing the performance of MLP.
"""
print("Tesing the performance of MLP....")
# Load data
data, label = load_boston_house_prices()
data = min_max_scale(data)
# Split data randomly, train set rate 70%
data_train, data_test, label_train, label_test = train_test_split(
data, label, random_state=20)
# Train model
reg = MLP()
reg.fit(data=data_train, label=label_train, n_hidden=8,
epochs=500, batch_size=8, learning_rate=0.0008)
# Model evaluation
get_r2(reg, data_test, label_test)
print(reg)