def deserialize_elastic_regressor(model_dict):
model = ElasticNet(model_dict["params"])
model.coef_ = np.array(model_dict["coef_"])
model.alpha = np.array(model_dict["alpha"]).astype(np.float)
if isinstance(model_dict["n_iter_"], list):
model.n_iter_ = np.array(model_dict["n_iter_"])
else:
model.n_iter_ = int(model_dict["n_iter_"])
if isinstance(model_dict["intercept_"], list):
model.intercept_ = np.array(model_dict["intercept_"])
else:
model.intercept_ = float(model_dict["intercept_"])
return model
def deserialize_elastic_regressor(model_dict):
model = ElasticNet(model_dict['params'])
model.coef_ = np.array(model_dict['coef_'])
model.alpha = np.array(model_dict['alpha']).astype(np.float)
if isinstance(model_dict['n_iter_'], list):
model.n_iter_ = np.array(model_dict['n_iter_'])
else:
model.n_iter_ = int(model_dict['n_iter_'])
if isinstance(model_dict['intercept_'], list):
model.intercept_ = np.array(model_dict['intercept_'])
else:
model.intercept_ = float(model_dict['intercept_'])
return model
######################################### # ELASTIC NET MODEL from sklearn.linear_model import ElasticNet regr = ElasticNet(random_state=0) regr.alpha = 1.9 regr.l1_ratio = 0.65 regr.fit(X_train,y_train) # Calling the score method, which compares the predicted values to the actual values y_score = regr.score(X_test, y_test) # The score is directly comparable to R-Square print(y_score) #########
label='Test data')
plt.xlabel('Predicted values')
plt.ylabel('Residuals')
plt.legend(loc='upper left')
plt.hlines(y=0, xmin=-10, xmax=50, color='black', lw=2)
plt.xlim([-10, 50])
plt.show()
# Setup the array of alphas and lists to store scores
alpha_space = np.logspace(-4, 0, 50)
MSE3_test_scores = []
MSE3_train_scores = []
R23_test_scores = []
R23_train_scores = []
for alpha in alpha_space:
sl.alpha = alpha
sl.fit(X, y)
y_train_pred3 = sl.predict(X_train)
y_test_pred3 = sl.predict(X_test)
MSE3_test_scores.append(mean_squared_error(y_test, y_test_pred3))
MSE3_train_scores.append(mean_squared_error(y_train, y_train_pred3))
R23_test_scores.append(r2_score(y_test, y_test_pred3))
R23_train_scores.append(r2_score(y_train, y_train_pred3))
plt.plot(alpha_space, MSE3_test_scores)
plt.xlabel('alpha_space')
plt.ylabel('MSE3_test_scores')
plt.show()
plt.plot(alpha_space, MSE3_train_scores)
plt.xlabel('alpha_space')
plt.ylabel('MSE3_train_scores')
plt.savefig('ElasticNet l1_ratio=' + str(alpha) + '10_09.png', dpi=300)
plt.show()
alpha_space = np.logspace(-4, 0, 50)
elanet_scores = []
elanet_scores_std = []
# Create a ridge regressor: ridge
#elanet = ElasticNet(normalize=True)
elanet = ElasticNet(alpha=1.0, l1_ratio=0.5, normalize=True)
# Compute scores over range of alphas
for alpha in alpha_space:
# Specify the alpha value to use: ridge.alpha
elanet.alpha = alpha
# Perform 10-fold CV: ridge_cv_scores
elanet_cv_scores = cross_val_score(elanet, X_train, y_train, cv=10)
# Append the mean of ridge_cv_scores to ridge_scores
elanet_scores.append(np.mean(elanet_cv_scores))
# Append the std of ridge_cv_scores to ridge_scores_std
elanet_scores_std.append(np.std(elanet_cv_scores))
# Display the plot
display_plot(elanet_scores, elanet_scores_std, 'elanet')
print("My name is Yuchen Duan")
print("My NetID is: yuchend3")
return mse
# %%
#To be looped for each fold
for i in range(5):
elastic_net = ElasticNet(l1_ratio=0.5)
coefs = pd.DataFrame(index=lambdas, columns=train[0].columns)
mse = pd.DataFrame(index=lambdas, columns=['MSE'])
sample = pd.DataFrame(np.concatenate(train[:i] + train[i + 1:], axis=0))
X = sample.cov()
Y = sample.mean(axis=0)
for j, l in enumerate(lambdas):
elastic_net.alpha = l # set the severity of the constraint
elastic_net.fit(X, Y)
coefs.iloc[j] = elastic_net.coef_
mse.iloc[j] = MSE(X, Y, elastic_net)
mse_summary = pd.concat([mse_summary, mse], axis=1)
# %%
minimum_mse = mse_summary.mean(axis=1).min()
optimal_lambda = mse_summary.mean(axis=1).idxmin()
#The minimum mse
print(minimum_mse, '\n')
#The lambda that gives minimum mse
print(optimal_lambda, '\n')
# %%