from sklearn.metrics import confusion_matrix

import matplotlib.pyplot as plt

import numpy as np

import os

import cv2

from keras.models import load_model

import sklearn.metrics as metrics

from SiameseResNet_test2 import building_block, create_model, create_CaptureFeature_model

from sklearn.preprocessing import label_binarize

import time

allow_path = './data/auc_mc2/allow'

reject_path = './data/auc_mc2/reject'

anchor_name = ['jeff', 'jiawei', 'donna', 'peng', 'fannie']

# anchor_name = ['chieh', 'jeff', 'jiawei', 'donna', 'peng']

nb_classes = len(anchor_name) + 1

y_true = []

y_pred = []

#----------- create y_true -------------

for allow in os.listdir(allow_path): # allow

for i in os.listdir(allow_path + '/' + allow):

y_true.append(anchor_name.index(allow))

for reject in os.listdir(reject_path): # reject

for i in os.listdir(reject_path + '/' + reject):

y_true.append(5)

y_true = np.array(y_true)

Y_true = label_binarize(y_true, classes=[i for i in range(nb_classes)])

print('k=', y_true.shape)

#----------- create y_pred -------------

ori_img_array = []

origin_model = load_model('./model_with5/SiameseResnet_mc2_model/SiameseResnet_mc2_stable_.h5')

fix_model = create_CaptureFeature_model((1, 32, 32))

fix_model.set_weights(origin_model.get_weights()) #.layers[3]

relation_model = load_model('./model_with5/test_by_test.h5')

#--- fix_feature ---

fix_img = np.load('./npz/fix_image_sum_np.npy')

fix_img = fix_img / 255.0

fix_feature = fix_model.predict(fix_img)

fix_feature = fix_feature[np.newaxis, :,:,:,:]

print(fix_feature.shape)

print(len(fix_img))

#--- ori_feature ---

for i in os.listdir(allow_path):

for j in os.listdir(allow_path + '/' + i):

ori_img = cv2.imread(allow_path + '/' + i + '/' + j, cv2.IMREAD_UNCHANGED)

ori_img_array.append(ori_img)

for i in os.listdir(reject_path):

for j in os.listdir(reject_path + '/' + i):

ori_img = cv2.imread(reject_path + '/' + i + '/' + j, cv2.IMREAD_UNCHANGED)

ori_img_array.append(ori_img)

ori_img_np = np.array(ori_img_array, dtype='float32')

ori_img_np = ori_img_np[:, np.newaxis,:,:]

ori_img_np = ori_img_np / 255.0

print(ori_img_np.shape)

ori_feature = fix_model.predict(ori_img_np)

ori_feature = ori_feature[np.newaxis,:,:,:,:]

print(ori_feature.shape)

#--- repeat ---

fix_feature_repeat = np.repeat(fix_feature, ori_feature.shape[1], axis=0)

ori_feature_repeat = np.repeat(ori_feature, fix_feature.shape[1], axis=0)

ori_feature_repeat = np.swapaxes(ori_feature_repeat, 0, 1) # 轉置

print(fix_feature_repeat.shape)

print(ori_feature_repeat.shape)

#--- concatenate ---

cat_feature = np.concatenate([fix_feature_repeat, ori_feature_repeat], axis=2)

input_feature = np.reshape(cat_feature, (-1, 64, 4, 4))

print(input_feature.shape)

score = relation_model.predict(input_feature)

score = np.reshape(score, (-1, len(fix_img)))

print(score)

for i in range(len(score)):

result = np.argmax(score[i])

guess = score[i][result]

if result >= len(anchor_name):

y_pred.append(len(anchor_name))

else:

y_pred.append(result)

# print(y_pred)

y_pred = np.array(y_pred)

Y_pred = label_binarize(y_pred, classes=[i for i in range(nb_classes)])

#------

fpr, tpr, thresholds = metrics.roc_curve(Y_true.ravel(), Y_pred.ravel())

auc = metrics.auc(fpr, tpr)

print('fpr:', fpr)

print('tpr:', tpr)

timestamp = time.strftime('_%m%d%H%M%S', time.localtime())

txt = open('./auc/roc_curve'+ timestamp +'.txt', 'w')

txt.write('AUC: %.3f' % auc)

print('AUC: %.3f' % auc)

for i in range(len(thresholds)):

txt.write('\nthresholds:' + str(thresholds[i]))

txt.write('\nfpr:' + str(fpr[i]))

txt.write('\ntpr:' + str(tpr[i]))

txt.write('\ncount:' + str(len(Y_true)))

txt.write('\nthresholds count:' + str(len(thresholds)))

txt.close()

# plt.title('Receiver Operating Characteristic')

# plt.plot(fpr, tpr, 'b', label = 'AUC = %0.2f' % roc_auc)

# plt.legend(loc = 'lower right')

# plt.plot([0, 1], [0, 1],'r--')

# plt.xlim([0, 1])

# plt.ylim([0, 1])

# plt.ylabel('True Positive Rate')

# plt.xlabel('False Positive Rate')

# plt.show()

plt.plot([0, 1], [0, 1], linestyle='--')

plt.plot(fpr, tpr, marker='.')

plt.ylabel('True Positive Rate')

plt.xlabel('False Positive Rate')

plt.savefig('./auc/roc_curve'+ timestamp +'.png')

plt.show()