def train(train_x, train_y):
forest = rf.training().trainforest('rf', train_x, train_y, 1000, accuracy_train_calculation = True)
#===========================================================================
importance = rf.training().importance(forest,np.shape(train_x)[1])
feature_list = list(train_x.columns.values)
print feature_list
importance["keyword"] = importance["Ranking"].map(lambda x : feature_list[x])
print importance
# rf.training().dependence(forest, train_x, feature_set)
# rf.training().dependence3d(forest, train_x, feature_set)
#===========================================================================
return forest
#--------------------------------------------------------- tl.set_color('b')
#------------------------------------------------------------------------------
#------------------------------------------------------------------------------
#------------------------------------------------------------- ax2 = ax1.twinx()
#---------- ax2.plot(df['Tree'], pd.rolling_std(df['Score'], window = 10), 'r-')
#-------------------------------------- ax2.set_ylabel('Rolling std', color='r')
#---------------------------------------------- for tl in ax2.get_yticklabels():
#--------------------------------------------------------- tl.set_color('r')
#-------------------------------------------------------------------- plt.show()
#------------------------------------------------------------------------------
#------------------------------------------------------------------------------
#------------------------------------------------------------------------------
#------------------------------------------------- stop = timeit.default_timer()
#--------------------------- print "The running takes %r min" %((stop-start)/60)
ff = RFclass.training()
tt = RFclass.test()
pp = Superplot.fancy()
figsize(9.5, 7)
df1 = pd.read_csv(
'//home/peng/git/Machine_learning_for_reliability_analysis/Test_1/Results/Ensemble/statistical_csv/bag_acc_10cv_100_4000.csv',
header=0)
df2 = pd.read_csv(
'//home/peng/git/Machine_learning_for_reliability_analysis/Test_1/Results/Ensemble/statistical_csv/bag_prec_10cv_100_4000.csv',
header=0)
print "This is the training set of field data."
print "The size of the data is "
print df.describe()
plt.plot(df1['tree_range'], df1['12'], label='Accuracy')
plt.plot(df1['tree_range'], df2['12'], label='Precision')
def test(test_x, test_y, train_model):
rf.test().testforest(test_x, test_y, train_model)
#------------------------------------------------------------------------------
#------------------------------------------------------------- ax2 = ax1.twinx()
#---------- ax2.plot(df['Tree'], pd.rolling_std(df['Score'], window = 10), 'r-')
#-------------------------------------- ax2.set_ylabel('Rolling std', color='r')
#---------------------------------------------- for tl in ax2.get_yticklabels():
#--------------------------------------------------------- tl.set_color('r')
#-------------------------------------------------------------------- plt.show()
#------------------------------------------------------------------------------
#------------------------------------------------------------------------------
#------------------------------------------------------------------------------
#------------------------------------------------- stop = timeit.default_timer()
#--------------------------- print "The running takes %r min" %((stop-start)/60)
ff = RFclass.training()
tt = RFclass.test()
pp = Superplot.fancy()
figsize(9.5,7)
df1 = pd.read_csv('//home/peng/git/Machine_learning_for_reliability_analysis/Test_1/Results/Ensemble/statistical_csv/bag_acc_10cv_100_4000.csv', header=0)
df2 = pd.read_csv('//home/peng/git/Machine_learning_for_reliability_analysis/Test_1/Results/Ensemble/statistical_csv/bag_prec_10cv_100_4000.csv', header=0)
print "This is the training set of field data."
print "The size of the data is "
print df.describe()
plt.plot(df1['tree_range'], df1['12'], label='Accuracy')
plt.plot(df1['tree_range'], df2['12'], label = 'Precision')
plt.legend(fontsize = 20)
plt.xticks(fontsize =20)
plt.yticks(fontsize =20)
plt.ylabel('Classification metrics', fontsize = 24)
def training(self):
bestmodel= RFC(n_estimators= 100).fit(self.train,self.trainlabel)
ff= RFclass.training()
ff.importance(bestmodel, 12, color = '#66cdaa', plot_std = False)
return bestmodel
def training(self):
bestmodel = RFC(n_estimators=100).fit(self.train, self.trainlabel)
ff = RFclass.training()
ff.importance(bestmodel, 12, color='#66cdaa', plot_std=False)
return bestmodel
def main():
start = timeit.default_timer()
df = pd.read_csv('/home/peng/new160half.csv', header=0)
# df['random_number']=np.random.random(size = 160)
# df_sort = df.sort(columns='random_number')
# df_sort.drop(['random_number'], inplace = True, axis = 1)
# df_sort.to_csv('new_random_160.csv', header = 0)
p = Preprocessdata.standardprocess()
# # df_2 = pd.read_csv('/home/peng/git/Machine_learning_for_reliability_analysis//Test_1/score_long_2features_rf.csv', header=0)
#------------------------------------------------------------------------------
train, trainlabel, test, testlabel = p.noscale(df, 0.9)
# train, trainlabel = p.noaction(df)
# # df_2 = pd.read_csv('/home/peng/git/Machine_learning_for_reliability_analysis/score50-500_2.csv', header=0)
ff = RFclass.training()
tt = RFclass.test()
feature_range = np.arange(12, 13, 1)
tree_range = np.arange(700, 701, 1)
#####################sensitivity for 10cv##############
#---------------------------------------------------------------- score = []
#------------------------------------------------------ for i in range(100):
# score.append(ff.trainman_sensitivity_CV('adb', train, trainlabel, tree_range, feature_range))
#-------------------------------------------------------------- # print score
#- df_raw_times = pd.DataFrame({'times':np.arange(1,101,1), 'scores':score})
#------------------------------------------------------------------------------
#----------------------- df_raw_times = ff.str_float(df_raw_times['scores'])
#------------------------------------------------------------------------------
#---------------------------------- df_acc_times = ff.accuracy(df_raw_times)
#-- df_acc_times.to_csv('adb_acc_10cv_f12_t700_100times.csv', header = True)
"""Just separate"""
#===========================================================================
# forest= ff.trainforest('ext', train, trainlabel,1900,9)
# y_pred = forest.predict(test)
# print metrics.precision_score(testlabel,y_pred)
# cm = metrics.confusion_matrix(testlabel, y_pred)
# tt.plot_confusion_matrix(cm)
#===========================================================================
"""the CART single tree"""
forest = ff.trainforest('cart', train, trainlabel, 20, 1)
y_pred = forest.predict(test)
print metrics.accuracy_score(testlabel, y_pred)
print metrics.precision_score(testlabel, y_pred)
cm = metrics.confusion_matrix(testlabel, y_pred)
tt.plot_confusion_matrix(cm)
#---------------------------------------------------------------- score = []
#------------------------------------------------------ for i in range(100):
#----------- forest = ff.trainforest('adb', train, trainlabel, 1450, 11)
#----------------------------------------- y_pred = forest.predict(test)
#--------------- score.append(metrics.accuracy_score(testlabel, y_pred))
#------------------------------------------------------------------------------
#--------- df=pd.DataFrame({'times': np.arange(1,101,1), 'acc_score':score})
#--------------- df.to_csv('adb_acc_63_f12_t1450_100times.csv', header=True)
#------------------------------------------------------- print df.describe()
#------------------------------------------------------------------------------
#------------------------------------------------------------------------------
#------------------------------------------------------------------------------
# df = pd.read_csv('/home/peng/git/Machine_learning_for_reliability_analysis/Test_1/adb_acc_63_f12_t1450_100times.csv', header=0)
#------------------------------------ plt.plot(df['times'], df['acc_score'])
# plt.xticks(np.arange(1,101,1),np.arange(1,101,1))
# plt.xlabel()
plt.show()
############################################################################################################################
#------------------------------------- df_66_33 = {'tree_tange': tree_range}
#---------------------------------------------- df_all = DataFrame(df_66_33)
# scores = ff.trainonlyfeat('bag', train, trainlabel, tree_range, feature_range)
# scores.to_csv('bag_100_4000_10times.csv', header=True)
# data = ff.train_repeat_forest_metrics('bag', train, trainlabel, test, testlabel, tree_range, feature_range, 10)
#---------------- data.to_csv('nnnnnn_crazy66_33_100_4000.csv', header=True)
#------------------------------------------------------------------------------
# data = ff.train_repeat_forest_metrics('adb', train, trainlabel, test, testlabel, tree_range, feature_range, 10)
#------------ data.to_csv('nnnnnnnnnn_crazy66_33_100_4000.csv', header=True)
# data = ff.train_repeat_forest_metrics('gbt', train, trainlabel, test, testlabel, tree_range, feature_range, 10)
#------------------- data.to_csv('gbt_crazy66_33_100_4000.csv', header=True)
#-- data = ff.trainmanCV('rf', train, trainlabel, tree_range, feature_range)
#-------------------------- data.to_csv('rf_crazy100_4000.csv', header=True)
#------------------------------------------------------------------------------
#- data = ff.trainmanCV('ext', train, trainlabel, tree_range, feature_range)
#------------------------- data.to_csv('ext_crazy100_4000.csv', header=True)
#- data = ff.trainmanCV('bag', train, trainlabel, tree_range, feature_range)
#------------------------ data.to_csv('bag_crazy100_4000n.csv', header=True)
#------------------------------------------------------------------------------
#- data = ff.trainmanCV('adb', train, trainlabel, tree_range, feature_range)
#------------------------ data.to_csv('adb_crazy100_4000n.csv', header=True)
#- data = ff.trainmanCV('gbt', train, trainlabel, tree_range, feature_range)
#------------------------ data.to_csv('gbt_crazy100_4000n.csv', header=True)
# scores.to_csv('rf_1_5_1_feature4.csv', header=False)
# print scores
stop = timeit.default_timer()
print "The running takes %r min" % ((stop - start) / 60)
def testRF(test_x, test_y, forest):
output_RF = rf.test().testforest(test_x, test_y, forest)
def trainRF(train_x, train_y, n_estimator):
forest = rf.training().trainforest('rf', train_x, train_y, n_estimator)
importances = rf.training().importance(forest, n_estimator)
return forest, importances
