#norm age and rating
#data[0:,-2] = data[0:,-2] / data[0:,-2].max()
#data[0:,-1] = data[0:,-1] / data[0:,-1].max()
#data_word_age = data[0:,0:-1]
train_x = data[0:, 0:-1]
train_y = np.array(data[0:, -1:]).reshape((data.shape[0], ))
X_train, X_test, y_train, y_test = train_test_split(train_x,
train_y,
test_size=0.3,
random_state=0)
reg = RandomForestRegressor()
reg.fit(X_train, y_train)
p = reg.predict(X_test)
s = reg.score(X_test, y_test)
print(s)
# above mean or not
train_rating_average = np.average(train_y)
binary_p = np.zeros([p.shape[0]])
iteration = 0
for i in p:
if i > train_rating_average:
binary_p[iteration] = 1.0
else:
binary_p[iteration] = 0.0
iteration += 1
binary_ytest = np.zeros([p.shape[0]])
iteration = 0
for i in y_test:
for column in features:
if plotnumber <= len(features):
ax = plt.subplot(4, 4, plotnumber)
sns.stripplot(target, features[column])
plotnumber += 1
plt.show()
# In[15]:
from sklearn.ensemble.forest import RandomForestRegressor
# In[16]:
rand_clf = RandomForestRegressor()
# In[272]:
x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.30)
# In[273]:
rand_clf.fit(x_train, y_train)
# In[274]:
rand_clf.score(x_train, y_train)
# In[275]:
rand_clf.score(x_test, y_test)
# Generate training and test set
X_train,y_train = OrganizeData(nucleus, 'train')
X_test, y_test = OrganizeData(nucleus, 'test')
# Feature scaling
X_train_scaled = preprocessing.scale(X_train)
X_test_scaled = preprocessing.scale(X_test)
# Set the parameters for the random forest estimator
estimator = RandomForestRegressor(n_estimators=50, max_features=16, max_depth=25,
min_samples_split=5, min_samples_leaf=5, random_state=0)
# Build the random forest of regression trees from the training set
estimator = estimator.fit(X_train_scaled,y_train)
print estimator.score(X_train_scaled,y_train)
print estimator.score(X_test_scaled,y_test)
# Predict regression target for the test set
predicted = estimator.predict(X_train_scaled)
cc = np.corrcoef(y_train,predicted)
print cc
print estimator
#my_plotting.simple_plot_overlay(y_train,predicted)
predicted = estimator.predict(X_test_scaled)
cc = np.corrcoef(y_test,predicted)
print cc
print estimator
#my_plotting.simple_plot_overlay(y_test,predicted)
# if res[i][j] == 100:
# res[i][j] = 0
# else:
# res[i][j] = -0.01 * res[i][j]
# return res
# def normalizeY(arr):
# arr=arr/100
# return arr
if __name__ == '__main__':
train_x, test_x, train_y, test_y, x_data, y_data = load(train_data_path)
rf_model = RandomForestRegressor()
rf_model.fit(x_data, y_data)
with open(filename, 'wb') as file:
pickle.dump(rf_model, file)
rf_train_score = rf_model.score(x_data, y_data)
rf_test_score = rf_model.score(test_x, test_y)
print("RF train score:",rf_train_score)
print("RF test score:",rf_test_score)
dt_model = DecisionTreeRegressor()
dt_model.fit(x_data, y_data)
with open(filename2, 'wb') as file:
pickle.dump(dt_model, file)
dt_train_score = dt_model.score(x_data, y_data)
dt_test_score = dt_model.score(test_x, test_y)
print("DT train score:",dt_train_score)
print("DT test score:",dt_test_score)
# fit a linear model with no bells and whistles
model = linear_model.LinearRegression()
model.fit(train_X, train_Y)
# look at the r squared on the training data and the test data
model.score(train_X, train_Y)
model.score(test_X, test_Y)
# See if I can get the r squared on the test data lower by using more complex models
# random forest
forest = RandomForestRegressor()
# fit the data without using cross val to select parameters
# note that train score is much higher than test score
forest.fit(train_X, train_Y)
forest.score(train_X, train_Y)
forest.score(test_X, test_Y)
# fit a random forest regressor using grid search to
# select the number of trees and max depth
new_forest = RandomForestRegressor()
params_grid = [{'max_depth': [3, 5,10, None], 'n_estimators': [5,10,15,20, 50, 80]} ]
grid_search = GridSearchCV(new_forest, params_grid, cv=10)
grid_search.fit(train_X, train_Y)
grid_search.score(test_X, test_Y)
grid_search.best_estimator_
# fit a boosted regression
boost = GradientBoostingRegressor()
params_grid = [{'learning_rate': [.05,.1,2], 'n_estimators': [20,50,100,150]} ]
