Python metrics.precision_score函数代码示例

def nearest_centroid(input_file,Output,test_size):
    ncol=tools.file_col_coma(input_file)
    data = np.loadtxt(input_file, delimiter=',', usecols=range(ncol-1))
    X = data[:,1:]
    y = data[:,0]
    n_samples, n_features = X.shape
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=test_size)
    print X_train.shape, X_test.shape
    clf = NearestCentroid()
    clf.fit(X_train,y_train)
    y_pred = clf.predict(X_test)
    print "Nearest Centroid Classifier "
    print "classification accuracy:", metrics.accuracy_score(y_test, y_pred)
    print "precision:", metrics.precision_score(y_test, y_pred)
    print "recall:", metrics.recall_score(y_test, y_pred)
    print "f1 score:", metrics.f1_score(y_test, y_pred)
    print "\n"
    results = Output+"Nearest_Centroid_metrics_test.txt"
    file = open(results, "w")
    file.write("Nearest Centroid Classifier estimator accuracy\n")
    file.write("Classification Accuracy Score: %f\n"%metrics.accuracy_score(y_test, y_pred))
    file.write("Precision Score: %f\n"%metrics.precision_score(y_test, y_pred))
    file.write("Recall Score: %f\n"%metrics.recall_score(y_test, y_pred))
    file.write("F1 Score: %f\n"%metrics.f1_score(y_test, y_pred))
    file.write("\n")
    file.write("True Value, Predicted Value, Iteration\n")
    for n in xrange(len(y_test)):
        file.write("%f,%f,%i\n"%(y_test[n],y_pred[n],(n+1)))
    file.close()
    title = "Nearest Centroid %f"%test_size
    save = Output + "Nearest_Centroid_confusion_matrix"+"_%s.png"%test_size
    plot_confusion_matrix(y_test, y_pred,title,save)
    lvltrace.lvltrace("LVLSortie dans stochasticGD split_test")

def gaussianNB(input_file,Output,test_size):
    lvltrace.lvltrace("LVLEntree dans gaussianNB split_test")
    ncol=tools.file_col_coma(input_file)
    data = np.loadtxt(input_file, delimiter=',', usecols=range(ncol-1))
    X = data[:,1:]
    y = data[:,0]
    n_samples, n_features = X.shape
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=test_size)
    print X_train.shape, X_test.shape
    # Instantiate the estimator
    clf = GaussianNB()
    # Fit the estimator to the data
    clf.fit(X_train, y_train)
    # Use the model to predict the last several labels
    y_pred = clf.predict(X_test)
    print "Gaussian Naive Bayes estimator accuracy "
    print "classification accuracy:", metrics.accuracy_score(y_test, y_pred)
    print "precision:", metrics.precision_score(y_test, y_pred)
    print "recall:", metrics.recall_score(y_test, y_pred)
    print "f1 score:", metrics.f1_score(y_test, y_pred)
    results = Output+"GaussianNB_metrics_test.txt"
    file = open(results, "w")
    file.write("Gaussian Naive Bayes estimator accuracy\n")
    file.write("Classification Accuracy Score: %f\n"%metrics.accuracy_score(y_test, y_pred))
    file.write("Precision Score: %f\n"%metrics.precision_score(y_test, y_pred))
    file.write("Recall Score: %f\n"%metrics.recall_score(y_test, y_pred))
    file.write("F1 Score: %f\n"%metrics.f1_score(y_test, y_pred))
    file.write("True Value, Predicted Value, Iteration\n")
    for n in xrange(len(y_test)):
        file.write("%f,%f,%i\n"%(y_test[n],y_pred[n],(n+1)))
    file.close()
    title = "Gaussian Naive Bayes %f"%test_size
    save = Output + "Gaussian_NB_confusion_matrix"+"_%s.png"%test_size
    plot_confusion_matrix(y_test, y_pred,title,save)
    lvltrace.lvltrace("LVLsortie dans gaussianNB split_test")

def predictSVD(svd, row, column, d):
    # start = timeit.default_timer()
    u = svd[0] #clf.components_ 
    s = svd[1] #clf.explained_variance_
    vt = svd[2] #clf.fit_transform(X)
    # print "   fitting done.";
    # stop = timeit.default_timer()
    # print "   runtime: " + str(stop - start)
    # print "d:"
    # print d

    # matrixY = clf.components_ 
    probsY = []
    # print "dot products:"
    for i in range(len(row)):
        # print np.dot(u[:,column[i]], v[row[i],:])
        prob = np.sum(u[column[i],:]*s*vt[:,row[i]])
        if(prob < 0): prob = 0
        if(prob > 1): prob = 1
        probsY.append(prob)

    probsY = np.array(probsY)
    preds = np.zeros(shape=len(probsY))
    preds[probsY >= 0.5] = 1

    print "Precision"
    print precision_score(d, preds)
    print "Recall"
    print recall_score(d, preds)
    print "F-Score"
    print f1_score(d, preds)

    return probsY, preds

def SVC_linear(input_file,Output,test_size):
    lvltrace.lvltrace("LVLEntree dans SVC_linear split_test")
    ncol=tools.file_col_coma(input_file)
    data = np.loadtxt(input_file, delimiter=',', usecols=range(ncol-1))
    X = data[:,1:]
    y = data[:,0]
    n_samples, n_features = X.shape
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=test_size)
    print X_train.shape, X_test.shape
    clf=svm.SVC(kernel='linear')
    clf.fit(X_train,y_train)
    y_pred = clf.predict(X_test)
    print "C-Support Vector Classifcation (with RBF linear) "
    print "y_test, y_pred, iteration"
    print "classification accuracy:", metrics.accuracy_score(y_test, y_pred)
    print "precision:", metrics.precision_score(y_test, y_pred)
    print "recall:", metrics.recall_score(y_test, y_pred)
    print "f1 score:", metrics.f1_score(y_test, y_pred)
    print "\n"
    results = Output+"SVM_Linear_Kernel_metrics_test.txt"
    file = open(results, "w")
    file.write("Support Vector Machine with Linear Kernel estimator accuracy\n")
    file.write("Classification Accuracy Score: %f\n"%metrics.accuracy_score(y_test, y_pred))
    file.write("Precision Score: %f\n"%metrics.precision_score(y_test, y_pred))
    file.write("Recall Score: %f\n"%metrics.recall_score(y_test, y_pred))
    file.write("F1 Score: %f\n"%metrics.f1_score(y_test, y_pred))
    file.write("\n")
    file.write("True Value, Predicted Value, Iteration\n")
    for n in xrange(len(y_test)):
        file.write("%f,%f,%i\n"%(y_test[n],y_pred[n],(n+1)))
    file.close()
    title = "SVC linear %f"%test_size
    save = Output + "SVC_linear_confusion_matrix"+"_%s.png"%test_size
    plot_confusion_matrix(y_test, y_pred,title,save)
    lvltrace.lvltrace("LVLsortie dans SVC_linear split_test")

def evaluate(ytest, ypred, filename='metrics.txt'):
    true_result = [1 if item > 0.5 else 0 for item in ytest]
    pred_result = [1 if item > 0.5 else 0 for item in ypred]
    
    cm = confusion_matrix(true_result, pred_result)
    print('\nConfusion matrix:')
    print(cm)
    print("\nLoss classified as loss", cm[0][0])
    print("Wins classified as wins", cm[1][1])
    print("Wins classified as loss", cm[1][0])
    print("Loss classified as wins", cm[0][1])
    print('\nAccuracy:\t', accuracy_score(true_result, pred_result))
    print('Precision:\t', precision_score(true_result, pred_result))
    print('Recall: \t', recall_score(true_result, pred_result))
    print('F1 score:\t', f1_score(true_result, pred_result))
    print('Mean absolute error:\t', mean_absolute_error(ytest, ypred))
    
    # print to file
    print("Loss classified as loss", cm[0][0], file=open(filename, "a"))
    print("Wins classified as wins", cm[1][1], file=open(filename, "a"))
    print("Wins classified as loss", cm[1][0], file=open(filename, "a"))
    print("Loss classified as wins", cm[0][1], file=open(filename, "a"))
    print('\nAccuracy:\t', accuracy_score(true_result, pred_result), file=open(filename, "a"))
    print('Precision:\t', precision_score(true_result, pred_result), file=open(filename, "a"))
    print('Recall: \t', recall_score(true_result, pred_result), file=open(filename, "a"))
    print('F1 score:\t', f1_score(true_result, pred_result), file=open(filename, "a"))
    print('Mean absolute error:\t', mean_absolute_error(ytest, ypred), file=open(filename, "a"))

    def trainModel(self,folds):
        
        kf = cross_validation.StratifiedKFold(self.y_total,n_folds=folds,shuffle=True,random_state=random.randint(1,100))

        for (train_index,test_index) in (kf):
          
            self.X_train = [self.X_total[i] for i in train_index]
            self.X_test = [self.X_total[i] for i in test_index] 
            self.y_train = [self.y_total[i] for i in train_index]
            self.y_test = [self.y_total[i] for i in test_index] 

            print "################"
            print "Original"
            print np.array(self.y_test)
            print "################"
            self.clf = self.clf.fit(self.X_train,self.y_train)
            print "Predicted"
            y_pred = self.clf.predict(self.X_test)
            print y_pred
            print "################"
            print "Evaluation\n"           
            cm = confusion_matrix(self.y_test,y_pred)            
            print cm
            print "Precision Score:"
            print precision_score(self.y_test,y_pred,average="macro")
            print "Recall Score:"
            print recall_score(self.y_test,y_pred,average="macro") 
            print "Accuracy Score:"
            print accuracy_score(self.y_test,y_pred)

def learning_curve_mod(data, labels, clf, percents, d=100, avg=3, test_size=.2):
    """
    This method calculates the performance of the training and cross validation test set as the training
    set size increases and returns the performance at each percent

    Args:
        :param data: (md.array) The raw data to use for training and cross validation testing
        :param labels: (nd.array) the labels associated with the data
        :param clf: (sklearn classifier) the classifier to be used for training
        :param percents: (nd.array) a list of percent of training data to use
        :param d:  (int) The number of principle components to calculate
        :param avg: (int) The number of iterations to average when calculating performance
        :param test_size: (double [0,1]) The size of the testing set

    Return:
        :return: train_accuracies (list) performance on the training set
        :return: test_accuracies (list) performance on the testing set
    """
    # split into train and testing dataset
    x_train, x_test, y_train, y_test = train_test_split(data.T, labels, test_size=test_size, random_state=0)
    x_test = x_test.T
    train_accuracies = []
    test_accuracies = []
    for percent in percents:
        temp_train_accuracies = []
        temp_test_accuracies = []
        print percent
        for i in range(0, avg):
            x_train_2, x_test_2, y_train_2, y_test_2 = train_test_split(x_train, y_train, test_size=percent)
            x_train_2 = x_train_2.T

            # Subtract off the mean
            mean_face = np.mean(x_train_2, axis=1)
            x_train_2 = x_train_2 - mean_face

            # Find low dimensional subspace using PCA
            pca = PCA(n_components=d)
            pca.fit(x_train_2)
            model = pca.transform(x_train_2)

            # Project the known faces onto the face space
            label_map = np.dot(x_train_2.T, model)

            # Train a KNN classifier
            clf.fit(label_map, y_train_2)

            # project the unknown faces onto face space
            W_train = np.dot(x_train_2.T - mean_face.T, model)
            W_test = np.dot(x_test.T - mean_face.T, model)


            test_prediction = clf.predict(W_test)
            temp_test_accuracies.append(metrics.precision_score(y_test, test_prediction))
            train_prediction = clf.predict(W_train)
            temp_train_accuracies.append(metrics.precision_score(y_train_2, train_prediction))

        train_accuracies.append(np.mean(temp_train_accuracies))
        test_accuracies.append(np.mean(temp_test_accuracies))

    return train_accuracies, test_accuracies

def main():
    resize_shape = 64
    print "data is loading..."
    train_X, train_Y, test_X, test_Y = load_data(resize_shape)
    print "data is loaded"
    print "feature engineering..."
    learning_rate = 0.01
    training_iters = 100000
    batch_size = 128
    display_step = 10

    # Network Parameters
    n_input = resize_shape*resize_shape # MNIST data input (img shape: 28*28)
    n_classes = 62 # MNIST total classes (0-9 digits)
    dropout = 0.5 # Dropout, probability to keep units

    with tf.Session() as sess:
        cnn = CNN(sess, learning_rate, training_iters, batch_size, display_step, n_input, n_classes, dropout,resize_shape)
        train_X = cnn.inference(train_X)
        test_X = cnn.inference(test_X)

    print "feature engineering is complete"

    print 'training phase'
    clf = svm.LinearSVC().fit(train_X, train_Y)
    print 'test phase'
    predicts = clf.predict(test_X)

    # measure function
    print 'measure phase'
    print confusion_matrix(test_Y, predicts)
    print f1_score(test_Y, predicts, average=None)
    print precision_score(test_Y, predicts, average=None)
    print recall_score(test_Y, predicts, average=None)
    print accuracy_score(test_Y, predicts)

def randomforest(input_file,Output):
    lvltrace.lvltrace("LVLEntree dans randomforest")
    ncol=tools.file_col_coma(input_file)
    data = np.loadtxt(input_file, delimiter=',', usecols=range(ncol-1))
    X = data[:,1:]
    y = data[:,0]
    n_samples, n_features = X.shape
    clf = RandomForestClassifier(n_estimators=10)
    clf.fit(X,y)
    y_pred = clf.predict(X)
    print "#########################################################################################################\n"
    print "The Random forest algo "
    print "classification accuracy:", metrics.accuracy_score(y, y_pred)
    print "precision:", metrics.precision_score(y, y_pred)
    print "recall:", metrics.recall_score(y, y_pred)
    print "f1 score:", metrics.f1_score(y, y_pred)
    print "\n"
    print "#########################################################################################################\n"
    results = Output+"Random_Forest_metrics.txt"
    file = open(results, "w")
    file.write("Random Forest Classifier estimator accuracy\n")
    file.write("Classification Accuracy Score: %f\n"%metrics.accuracy_score(y, y_pred))
    file.write("Precision Score: %f\n"%metrics.precision_score(y, y_pred))
    file.write("Recall Score: %f\n"%metrics.recall_score(y, y_pred))
    file.write("F1 Score: %f\n"%metrics.f1_score(y, y_pred))
    file.write("\n")
    file.write("True Value, Predicted Value, Iteration\n")
    for n in xrange(len(y)):
        file.write("%f,%f,%i\n"%(y[n],y_pred[n],(n+1)))
    file.close()
    title = "The Random forest"
    save = Output + "Random_Forest_confusion_matrix.png"
    plot_confusion_matrix(y, y_pred,title,save)
    lvltrace.lvltrace("LVLSortie dans randomforest")

def _clf_mlp(trX,teX,trY,teY):
	print "MLP"
	print trX.shape,"trX shape"
	print "Enter Layer for MLP"
	layer=input()
	# print "enter delIdx"
	# delIdx=input()
	# while(delIdx):
	# 	trX=np.delete(trX,-1,axis=0)
	# 	trY=np.delete(trY,-1,axis=0)
	# 	delIdx=delIdx-1
	print "factors",factors(trX.shape[0])	
	teY=teY.astype(np.int32)
	trY=trY.astype(np.int32)
	print trX.shape,"trX shape"
	print "enter no of mini batch"
	mini_batch=int(input())
	mlp = TfMultiLayerPerceptron(eta=0.01, 
                             epochs=100, 
                             hidden_layers=layer,
                             activations=['relu' for i in range(len(layer))],
                             print_progress=3, 
                             minibatches=mini_batch, 
                             optimizer='adam',
                             random_seed=1)
	mlp.fit(trX,trY)
	pred=mlp.predict(teX)
	print _f_count(teY),"test f count"
	pred=pred.astype(np.int32)
	print _f_count(pred),"pred f count"
	conf_mat=confusion_matrix(teY, pred)
	process_cm(conf_mat, to_print=True)
	print precision_score(teY,pred),"Precision Score"
	print recall_score(teY,pred),"Recall Score"
	print roc_auc_score(teY,pred), "ROC_AUC"

def stochasticGD(input_file,Output):
    lvltrace.lvltrace("LVLEntree dans stochasticGD")
    ncol=tools.file_col_coma(input_file)
    data = np.loadtxt(input_file, delimiter=',', usecols=range(ncol-1))
    X = data[:,1:]
    y = data[:,0]
    n_samples, n_features = X.shape
    clf = SGDClassifier(loss="hinge", penalty="l2")
    clf.fit(X,y)
    y_pred = clf.predict(X)
    print "#########################################################################################################\n"
    print "Stochastic Gradient Descent "
    print "classification accuracy:", metrics.accuracy_score(y, y_pred)
    print "precision:", metrics.precision_score(y, y_pred)
    print "recall:", metrics.recall_score(y, y_pred)
    print "f1 score:", metrics.f1_score(y, y_pred)
    print "\n"
    print "#########################################################################################################\n"
    results = Output+"Stochastic_GD_metrics.txt"
    file = open(results, "w")
    file.write("Stochastic Gradient Descent estimator accuracy\n")
    file.write("Classification Accuracy Score: %f\n"%metrics.accuracy_score(y, y_pred))
    file.write("Precision Score: %f\n"%metrics.precision_score(y, y_pred))
    file.write("Recall Score: %f\n"%metrics.recall_score(y, y_pred))
    file.write("F1 Score: %f\n"%metrics.f1_score(y, y_pred))
    file.write("\n")
    file.write("True Value, Predicted Value, Iteration\n")
    for n in xrange(len(y)):
        file.write("%f,%f,%i\n"%(y[n],y_pred[n],(n+1)))
    file.close()
    title = "Stochastic Gradient Descent"
    save = Output + "Stochastic_GD_confusion_matrix.png"
    plot_confusion_matrix(y, y_pred,title,save)
    lvltrace.lvltrace("LVLSortie dans stochasticGD")

def SVC_linear(input_file,Output):
    lvltrace.lvltrace("LVLEntree dans SVC_linear")
    ncol=tools.file_col_coma(input_file)
    data = np.loadtxt(input_file, delimiter=',', usecols=range(ncol-1))
    X = data[:,1:]
    y = data[:,0]
    n_samples, n_features = X.shape
    clf=svm.SVC(kernel='linear')
    clf.fit(X,y)
    y_pred = clf.predict(X)
    print "#########################################################################################################\n"
    print "C-Support Vector Classifcation (with linear kernel) "
    print "classification accuracy:", metrics.accuracy_score(y, y_pred)
    print "precision:", metrics.precision_score(y, y_pred)
    print "recall:", metrics.recall_score(y, y_pred)
    print "f1 score:", metrics.f1_score(y, y_pred)
    print "\n"
    print "#########################################################################################################\n"
    results = Output+"SVM_Linear_Kernel_metrics.txt"
    file = open(results, "w")
    file.write("Support Vector Machine with Linear Kernel estimator accuracy\n")
    file.write("Classification Accuracy Score: %f\n"%metrics.accuracy_score(y, y_pred))
    file.write("Precision Score: %f\n"%metrics.precision_score(y, y_pred))
    file.write("Recall Score: %f\n"%metrics.recall_score(y, y_pred))
    file.write("F1 Score: %f\n"%metrics.f1_score(y, y_pred))
    file.write("\n")
    file.write("True Value, Predicted Value, Iteration\n")
    for n in xrange(len(y)):
        file.write("%f,%f,%i\n"%(y[n],y_pred[n],(n+1)))
    file.close()
    title = "SVC - linear Kernel"
    save = Output + "SVC_linear_confusion_matrix.png"
    plot_confusion_matrix(y, y_pred,title,save)
    lvltrace.lvltrace("LVLSortie dans SVC_linear")

def run_model(X_test, X_train, y_test, y_train, prob_threshold = 20, layers = 5, nodes = 64, dropout = 50):
    
    print "run_model RUNNING"
    # Grab the model 
    model = get_model(X_test, layers =layers, dropout = dropout)
    model.fit(X_train, y_train, nb_epoch=20, batch_size=16, verbose = 0)

    # Get the training and test predictions from our model fit. 
    train_predictions  = model.predict_proba(X_train)
    test_predictions = model.predict_proba(X_test)
    # Set these to either 0 or 1 based off the probability threshold we 
    # passed in (divide by 100 becuase we passed in intergers). 
    train_preds = (train_predictions) >= prob_threshold / 100.0
    test_preds = (test_predictions) >= prob_threshold / 100.0

    # Calculate the precision and recall. Only output until 
    precision_score_train = precision_score(y_train, train_preds)
    precision_score_test = precision_score(y_test, test_preds)
    acc_train = accuracy_score(y_train, train_preds)
    acc_test = accuracy_score(y_test, test_preds)

    recall_score_train = recall_score(y_train, train_preds)
    recall_score_test = recall_score(y_test, test_preds)

    return precision_score_train, precision_score_test, recall_score_train, recall_score_test, acc_train, acc_test, model

def score(y_true, y_pred):
    precision_weighted = metrics.precision_score(
        y_true, y_pred, average='weighted')
    precision_ave = np.mean(metrics.precision_score(
        y_true, y_pred, average=None)[::12])

    recall_weighted = metrics.recall_score(
        y_true, y_pred, average='weighted')
    recall_ave = np.mean(metrics.recall_score(
        y_true, y_pred, average=None)[::12])

    f1_weighted = metrics.f1_score(
        y_true, y_pred, average='weighted')
    f1_ave = np.mean(metrics.f1_score(
        y_true, y_pred, average=None)[::12])

    stat_line = "  Precision: %0.4f\t Recall: %0.4f\tf1: %0.4f"
    res1 = "Weighted: " + stat_line % (100*precision_weighted,
                                       100*recall_weighted,
                                       100*f1_weighted)

    res2 = "Averaged: " + stat_line % (100*precision_ave,
                                       100*recall_ave,
                                       100*f1_ave)
    res3 = "-"*72
    outputs = [res3, res1, res2, res3]
    return "\n".join(outputs)

	def do_cv(self):
		from sklearn.metrics import precision_score


		data  = util.get_x_y_for_cv()
		X = data['x']
		y = data['y']['Survived']
		skf = StratifiedKFold(y, n_folds=5)
		e_test = []
		e_train = []
		for train_idx, test_idx in skf:
			train_x = X.iloc[train_idx]
			train_y = y.iloc[train_idx]

			test_x = X.iloc[test_idx]
			test_y = y.iloc[test_idx]

			self.model.fit(train_x, train_y)
			yhat = self.model.predict(test_x)
			e_test.append(precision_score(test_y, yhat))



			yhat_train = self.model.predict(train_x)
			e_train.append(precision_score(train_y, yhat_train))


		print np.mean(e_train)
		print np.mean(e_test)