def main():
outputdir = 'result/special_v3/'
if os.path.isdir(outputdir):
print('save in :' + outputdir)
else:
os.makedirs(outputdir)
train_img_path = '/data/MIT-BIH_AD_v3/'
test_img_path = '/data/MIT-BIH_AD_v3/'
train_file = './MIT-BIH_AD_sp_train.txt'
test_file = './MIT-BIH_AD_sp_val.txt'
num_classes = 7
f1 = open(train_file, 'r')
f2 = open(test_file, 'r')
lines = f1.readlines()
f1.close()
train_samples = len(lines)
lines = f2.readlines()
f2.close()
val_samples = len(lines)
batch_size = 32
epochs = 120
input_h = 128
input_w = 128
model = proposed_model(nb_classes=num_classes)
lr = 0.0001
adam = Adam(lr=lr)
model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=['accuracy'])
model.summary()
history = model.fit_generator(
generator_train_batch(train_file, batch_size, num_classes,
train_img_path, input_h, input_w),
steps_per_epoch=train_samples // batch_size,
epochs=epochs,
callbacks=[Step()],
validation_data=generator_val_batch(test_file, batch_size, num_classes,
test_img_path, input_h, input_w),
validation_steps=val_samples // batch_size,
verbose=1)
plot_history(history, outputdir)
save_history(history, outputdir)
model.save_weights(outputdir + 'proposed_model')
def main():
proposed = False
if proposed:
outputdir = 'result/NoAugment_{}/'.format(proposed)
if os.path.isdir(outputdir):
print('save in :'+outputdir)
else:
os.makedirs(outputdir)
train_img_path = '/data/MIT-BIH_AD/'
train_file = '/home/ccl/Documents/ECG-Arrhythmia-classification-in-2D-CNN/MIT-BIH_AD_train_paper.txt'
num_classes = 8
k = 10
f1 = open(train_file, 'r')
lines = f1.readlines()
f1.close()
train_samples = len(lines)
val_samples = len(lines)//k
num = len(lines)
new_lines = []
index = [n for n in range(num)]
random.shuffle(index)
for m in range(num):
new_lines.append(lines[index[m]])
lines = new_lines
temp = []
new_lines = []
for i in range(num):
if i % val_samples == 0:
temp = []
new_lines.append(temp)
temp.append(lines[i])
batch_size = 32
epochs = 40
input_h = 96
input_w = 96
augmentation = False
model = proposed_model()
lr = 0.0001
adam = Adam(lr=lr)
model.compile(loss='categorical_crossentropy', optimizer=adam, metrics=['accuracy'])
model.summary()
history = model.fit_generator(generator_train_batch_proposed(new_lines, k, batch_size, num_classes, train_img_path, input_h, input_w, augmentation=augmentation),
steps_per_epoch=train_samples // batch_size,
epochs=epochs,
callbacks=[Step()],
validation_data=generator_val_batch_proposed(new_lines, k, batch_size, num_classes, train_img_path, input_h, input_w, augmentation=augmentation),
validation_steps=val_samples // batch_size,
verbose=1)
plot_history(history, outputdir)
save_history(history, outputdir)
model.save_weights(outputdir+'proposed_model_{}.h5'.format(proposed))
else:
outputdir = 'result/NoAugment_{}/'.format(proposed)
if os.path.isdir(outputdir):
print('save in :' + outputdir)
else:
os.makedirs(outputdir)
train_img_path = '/data/MIT-BIH_AD/'
test_img_path = '/data/MIT-BIH_AD/'
train_file = '/home/ccl/Documents/ECG-Arrhythmia-classification-in-2D-CNN/MIT-BIH_AD_train.txt'
test_file = '/home/ccl/Documents/ECG-Arrhythmia-classification-in-2D-CNN/MIT-BIH_AD_val.txt'
num_classes = 8
f1 = open(train_file, 'r')
f2 = open(test_file, 'r')
lines = f1.readlines()
f1.close()
train_samples = len(lines)
lines = f2.readlines()
f2.close()
val_samples = len(lines)
batch_size = 32
epochs = 40
input_h = 96
input_w = 96
model = proposed_model()
lr = 0.0001
adam = Adam(lr=lr)
model.compile(loss='categorical_crossentropy', optimizer=adam, metrics=['accuracy'])
model.summary()
history = model.fit_generator(
generator_train_batch(train_file, batch_size, num_classes, train_img_path, input_h, input_w),
steps_per_epoch=train_samples // batch_size,
epochs=epochs,
callbacks=[Step()],
validation_data=generator_val_batch(test_file, batch_size, num_classes, test_img_path, input_h, input_w),
validation_steps=val_samples // batch_size,
verbose=1)
plot_history(history, outputdir)
save_history(history, outputdir)
model.save_weights(outputdir+'proposed_model_{}.h5'.format(proposed))
def main():
# init model
class_names = ['Normal', 'LBBB', 'RBBB', 'APC', 'PVC', 'PAB', 'VEB', 'VFW']
#PE is PAB
imageh = 128
imagew = 128
inputH = 96
inputW = 96
# ---------------------------change file models & weights--------------------
model = proposed_model()
lr = 0.0001
adm = Adam(lr=lr)
model.compile(loss='categorical_crossentropy', optimizer=adm, metrics=['accuracy'])
model.summary()
model.load_weights('result/first_attempt_False/proposed_model_False.h5', by_name=True)
# ---------------------------change models & weights--------------------
test_file = './MIT-BIH_AD_test.txt'
test_img_path = '/home/cc_lee/Dataset/MIT-BIH_AD'
augmentation = False
output_img = False
outputdir = os.path.join('./inference/', str(augmentation))
os.makedirs(outputdir, exist_ok=True)
os.makedirs(outputdir+'/False', exist_ok=True)
os.makedirs(outputdir+'/True', exist_ok=True)
f = open(test_file, 'r')
lines = f.readlines()
random.shuffle(lines)
TP = 0
count = 0
total = len(lines)
counter = {'Normal': 0, 'LBBB': 0, 'RBBB': 0, 'APC': 0, 'PVC': 0, 'PAB': 0, 'VEB': 0, 'VFW': 0}
tp_counter = {'Normal': 0, 'LBBB': 0, 'RBBB': 0, 'APC': 0, 'PVC': 0, 'PAB': 0, 'VEB': 0, 'VFW': 0}
for line in tqdm(lines):
path = line.split(' ')[0]
label = line.split(' ')[-1]
label = label.strip('\n')
answer = int(label)
img = os.path.join(test_img_path, path)
image = cv2.imread(img)
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
if augmentation:
Hshmean = int(np.round(np.max([0, np.round((imageh - inputH) / 2)])))
Wshmean = int(np.round(np.max([0, np.round((imagew - inputW) / 2)])))
image = image[Hshmean:Hshmean + inputH, Wshmean:Wshmean + inputW, :]
image = cv2.resize(image, (imagew, imageh))
else:
pass
input_data = np.zeros((1, imagew, imageh, 3), dtype='float32')
input_data[0] = image
pred = model.predict(input_data)
label = np.argmax(pred[0])
if label == answer:
TP += 1
tp_counter[class_names[label]] += 1
count += 1
counter[class_names[label]] += 1
if output_img:
if np.argmax(pred[0]) == 1:
color_t = (0, 255, 255)
else:
color_t = (0, 255, 0)
image = cv2.resize(image, (128*3, 128*3))
cv2.putText(image, class_names[answer].split(' ')[-1].strip(), (10, 30),
cv2.FONT_ITALIC, 1,
color_t, 1)
cv2.putText(image, class_names[label].split(' ')[-1].strip(), (10, 70),
cv2.FONT_HERSHEY_SIMPLEX, 1,
color_t, 1)
cv2.putText(image, "prob: %.4f" % pred[0][label], (10, 110),
cv2.FONT_HERSHEY_SIMPLEX, 1,
color_t, 1)
cv2.imwrite(os.path.join(outputdir, str(answer==label)) + '/' + '{}_{}'.format(class_names[answer], os.path.split(path)[1][:-4] + '_result.jpg', ), image)
print('{}/{} Acc: {} Pred:{} Answer: {}'.format(count, total, str(TP / count), class_names[label], class_names[answer] ) )
print('Normal:{}/{}={},\n LBBB:{}/{}={},\n RBBB:{}/{}={},\n APC:{}/{}={},\n PVC:{}/{}={},\n PAB:{}/{}={},\n VEB:{}/{}={},\n VFW:{}/{}={}'.format(
tp_counter['Normal'], counter['Normal'], (tp_counter['Normal']/counter['Normal']),
tp_counter['LBBB'], counter['LBBB'], (tp_counter['LBBB'] / counter['LBBB']),
tp_counter['RBBB'], counter['RBBB'], (tp_counter['RBBB'] / counter['RBBB']),
tp_counter['APC'], counter['APC'], (tp_counter['APC'] / counter['APC']),
tp_counter['PVC'], counter['PVC'], (tp_counter['PVC'] / counter['PVC']),
tp_counter['PAB'], counter['PAB'], (tp_counter['PAB'] / counter['PAB']),
tp_counter['VEB'], counter['VEB'], (tp_counter['VEB'] / counter['VEB']),
tp_counter['VFW'], counter['VFW'], (tp_counter['VFW'] / counter['VFW'])
))
def main():
outputdir = 'result/192_128_class_weight_v3_120eps/'
if os.path.isdir(outputdir):
print('save in :' + outputdir)
else:
os.makedirs(outputdir)
train_img_path = '/data/MIT-BIH_AD_v3/'
test_img_path = '/data/MIT-BIH_AD_v3/'
train_file = './MIT-BIH_AD_train.txt'
test_file = './MIT-BIH_AD_val.txt'
num_classes = 8
f1 = open(train_file, 'r')
f2 = open(test_file, 'r')
lines = f1.readlines()
f1.close()
train_samples = len(lines)
lines = f2.readlines()
f2.close()
val_samples = len(lines)
batch_size = 32
epochs = 120
input_h = 128
input_w = 128
class_weight = {
0: (1 - (75016 / 107620)) * 8,
1: (1 - (8072 / 107620)) * 8,
2: (1 - (7256 / 107620)) * 8,
3: (1 - (2544 / 107620)) * 8,
4: (1 - (7130 / 107620)) * 8,
5: (1 - (7024 / 107620)) * 8,
6: (1 - (106 / 107620)) * 8,
7: (1 - (472 / 107620)) * 8
}
model = proposed_model(nb_classes=num_classes)
lr = 0.0001
adam = Adam(lr=lr)
model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=['accuracy'])
model.summary()
history = model.fit_generator(
generator_train_batch(train_file, batch_size, num_classes,
train_img_path, input_h, input_w),
steps_per_epoch=train_samples // batch_size,
epochs=epochs,
callbacks=[Step()],
validation_data=generator_val_batch(test_file, batch_size, num_classes,
test_img_path, input_h, input_w),
validation_steps=val_samples // batch_size,
verbose=1,
class_weight=class_weight)
plot_history(history, outputdir)
save_history(history, outputdir)
model.save_weights(outputdir + 'proposed_model.h5')
