opdracht3versie2.py

# -*- coding: utf-8 -*-
"""
Created on Mon Feb 18 22:34:20 2019

@author: s164799
"""

from keras.models import model_from_json
import os

import numpy as np
import matplotlib.pyplot as plt

from keras.preprocessing.image import ImageDataGenerator
from keras.models import Sequential
from keras.layers import Dense, Flatten
from keras.layers import Conv2D, MaxPool2D
from keras.optimizers import SGD
from keras.callbacks import ModelCheckpoint, TensorBoard

# unused for now, to be used for ROC analysis
from sklearn.metrics import roc_curve, auc

TEST_PATH = 'C:/Users/s164799/Documents/jaar 3/kwartiel 3/Project imaging/test/'
MODEL_FILEPATH = 'my_first_cnn_model.json'
MODEL_WEIGHTS_FILEPATH = 'my_first_cnn_model_weights.hdf5'

json_file = open(MODEL_FILEPATH, 'r')
loaded_model_json = json_file.read()
json_file.close()
model = model_from_json(loaded_model_json)

model.load_weights(MODEL_WEIGHTS_FILEPATH)

def get_pcam_generators(base_dir, train_batch_size=32, val_batch_size=32):

     # dataset parameters
     TRAIN_PATH = os.path.join(base_dir, 'train+val', 'train')
     VALID_PATH = os.path.join(base_dir, 'train+val', 'valid')

     RESCALING_FACTOR = 1./255
     
     # instantiate data generators
     datagen = ImageDataGenerator(rescale=RESCALING_FACTOR)

     train_gen = datagen.flow_from_directory(TRAIN_PATH,
                                             target_size=(IMAGE_SIZE, IMAGE_SIZE),
                                             batch_size=train_batch_size,
                                             class_mode='binary')

     val_gen = datagen.flow_from_directory(VALID_PATH,
                                             target_size=(IMAGE_SIZE, IMAGE_SIZE),
                                             batch_size=val_batch_size,
                                             class_mode='binary',
                                             shuffle=False)
     
     return train_gen, val_gen
 
train_gen, val_gen = get_pcam_generators('/Users/s164799/Documents/jaar 3/kwartiel 3/Project imaging')
val_steps = val_gen.n//val_gen.batch_size  
predictions=model.predict_generator(val_gen,steps=val_steps)
y_pred = np.rint(predictions)
y_true = val_gen.classes

fpr_keras, tpr_keras, thresholds_keras = roc_curve(y_true, y_pred)
auc_keras = auc(fpr_keras, tpr_keras)

plt.figure()
plt.plot(fpr_keras, tpr_keras, color='darkorange', label='ROC curve (area = %0.2f)' % auc_keras)
plt.plot([0, 1], [0, 1], color='navy', lw=2, linestyle='--')
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('Receiver operating characteristic')
plt.legend(loc="lower right")
plt.show()