def Model_Structure_click(self):
self.get_current_path()
net, device = model_build(resume=True)
for i in net._modules.items():
self.textBrowser_image_3.append(str(i))
self.textBrowser_image_3.append(' ')
# 定义总参数量、可训练参数量及非可训练参数量变量
Total_params = 0
Trainable_params = 0
NonTrainable_params = 0
# 遍历model.parameters()返回的全局参数列表
for param in net.parameters():
mulValue = np.prod(param.size()) # 使用numpy prod接口计算参数数组所有元素之积
Total_params += mulValue # 总参数量
if param.requires_grad:
Trainable_params += mulValue # 可训练参数量
else:
NonTrainable_params += mulValue # 非可训练参数量
self.textBrowser_image_3.append('Total params: ' +
str(Total_params))
self.textBrowser_image_3.append('Trainable params: ' +
str(Trainable_params))
self.textBrowser_image_3.append('NonTrainable params: ' +
str(NonTrainable_params))
self.textBrowser_image_3.moveCursor(
self.textBrowser_image_3.textCursor().End) # 文本框顯示到底部
def predict(vocab_size, char2idx, idx2char):
model = model_build(vocab_size, embedding_dim, rnn_units, batch_size=1)
model.load_weights(tf.train.latest_checkpoint(checkpoint_dir))
model.build(tf.TensorShape([1, None]))
print(generate_text(model, start_string='星 '))
def train(dataset, vocab_size):
model = model_build(vocab_size=vocab_size, embedding_dim=embedding_dim, rnn_units=rnn_units, batch_size=BATCH_SIZE)
# 配置训练
model.compile(optimizer='adam', loss=loss)
checkpoint_callback = tf.keras.callbacks.ModelCheckpoint(filepath=checkpoint_prefix,
save_weights_only=True)
model.fit(dataset, epochs=EPOCHS, callbacks=[checkpoint_callback])
def Test_click(self):
self.get_current_path()
# print('self.index= ' + str(self.index))
transform_test = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.4914, 0.4822, 0.4465),
(0.2023, 0.1994, 0.2010)),
])
testset = datasets.CIFAR10(root='./data',
train=False,
download=True,
transform=transform_test)
# print('testset= ' + str(len(testset)))
testloader = DataLoader(testset,
batch_size=100,
shuffle=False,
num_workers=2)
# testloader = DataLoader(testset, batch_size=100, sampler=None, shuffle=False, num_workers=2)
# print('testloader= ' + str(len(testloader)))
classes_dict = {
0: 'airplane',
1: 'automobile',
2: 'bird',
3: 'cat',
4: 'deer',
5: 'dog',
6: 'frog',
7: 'horse',
8: 'ship',
9: 'truck'
}
net, device = model_build(resume=True)
predict(net, testloader, device, classes_dict, self.index)
img = cv2.imread(self.current_path + "/predict.png")
# print(img.shape)
cropped = img[0:1000, 240:730] # 裁剪坐标为[y0:y1, x0:x1]
img = cv2.resize(cropped, (331, 621))
io.imsave(self.current_path + "/predictPixmap.png", img)
pix = QPixmap(self.current_path + "/predictPixmap.png")
self.label_inputimage_17.setPixmap(pix)
DIGITS = 4
REVERSE = True
BATCH_SIZE = 128
# Maximum length of input is 'int + int' (e.g., '345+678'). Maximum length of
# int is DIGITS.
MAXLEN = DIGITS + 1 + DIGITS
# All the numbers, plus sign and space for padding.
chars = '0123456789+ '
ctable = CharacterTable(chars)
x_train, y_train, x_val, y_val = read_additions(DIGITS, MAXLEN, chars, ctable)
print('Build model...')
model = model_build(DIGITS, MAXLEN, chars)
model.summary()
# Train the model each generation and show predictions against the validation
# dataset.
for iteration in range(1, 20):
print()
print('-' * 50)
print('Iteration', iteration)
model.fit(x_train,
y_train,
batch_size=BATCH_SIZE,
epochs=1,
validation_data=(x_val, y_val))
# Select 10 samples from the validation set at random so we can visualize
# errors.
# Parameters for the model and dataset.
TRAINING_SIZE = 400000
DIGITS = 4
REVERSE = True
BATCH_SIZE = 128
# Maximum length of input is 'int + int' (e.g., '345+678'). Maximum length of
# int is DIGITS.
MAXLEN = DIGITS + 1 + DIGITS
# All the numbers, plus sign and space for padding.
chars = '0123456789+ '
ctable = CharacterTable(chars)
print('Load model...')
checkpoint = "addition_model.hdf5"
model = model_build(DIGITS, MAXLEN, chars, checkpoint=checkpoint)
model.summary()
# Predict addition provided by user
while(True):
print('-' * 50)
print('ATTENTION: ONLY {0} DIGITS MAX FOR EACH NUMBER, EXAMPLE: 4444+10'.format(DIGITS))
sentence = input('sentence: ')
_sentence = sentence + ' ' * (MAXLEN - len(sentence))
_sentence = _sentence[::-1]
x = ctable.encode(_sentence, MAXLEN)
preds = model.predict_classes(np.array([x]), verbose=0)
guess = ctable.decode(preds[0], calc_argmax=False)
print(sentence + ' = ' + guess)
when = str(args[3])
x = pd.read_csv(path + '/' + when + '_features_' + file)
y = pd.read_csv(path + '/' + when + '_target_'+ file)\
['Survived'].values
print('FILES NAME')
print('----> ' + path + '/' + when + '_features_' + file + ' <----')
print('----> ' + path + '/' + when + '_target_' + file + ' <----')
features = list(x.columns)
cat = to_list_categorical(x, do_print=False)
num = filter_list(raw=features, nasty=cat)
sectionizer('MODELING ', 65)
#----------------------------------------------------------------------
model = model_build(categorical_feat=cat, numerical_feat=num)
sectionizer(' - Train/Test Spliting ', 65)
#----------------------------------------------------------------------
x[num] = x[num].astype(float)
x_train, x_test,\
y_train, y_test = train_test_split(x,y,
test_size = 0.25,
random_state = 26)
sectionizer(' -- Training Model ', 65)
#----------------------------------------------------------------------
model.fit(x_train, y_train)
sectionizer('EVALUATING ', 65)
y_prob = model.predict_proba(x_test)
#Импорт пакетов и зависимостей import mnist #Набор данных MNIST #Загрузка набора данных train_images = mnist.train_images() #Изображения train-данных train_labels = mnist.train_labels() #Метки train-данных test_images = mnist.test_images() #Изображения test-данных test_labels = mnist.test_labels() #Метки test-данных import train train_images, test_images = train.reshape_images(train_images, test_images) import model model = model.model_build(train_images, train_labels, test_images, test_labels) import inference inference.test_show(model, test_images, test_labels, 9500, 9505)
wikidata = data.Dataset(params)
wikidata.read(wikipath)
print('wiki loaded')
traindata = data.Dataset(params)
traindata.read(trainpath, wikidata.char_id)
testdata = data.Dataset(params)
testdata.read(testpath, wikidata.char_id)
webdata = data.Dataset(params)
webdata.read(webpath, wikidata.char_id)
print('data loaded')
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
model = model.Model(params)
model.model_build(sess)
print('model built')
model.model_initialize(sess)
print('model initialize')
# model.lattice_train(wikidata, sess, 300000)
# print('wiki_lattice finished')
# model.train(traindata, sess, 40000)
# print('all_train finished')
# model.evaluate(testdata, sess)
# model.train(traindata, sess, 40000)
# print('train finished')
# model.evaluate(testdata, sess)
# print('lattice and text at the same time')
from tensorflow import keras
from model import model_build
# load data
fashion_mnist = keras.datasets.fashion_mnist
(train_images, train_labels), (test_images,
test_labels) = fashion_mnist.load_data()
train_images = train_images / 255.0
# train
model = model_build()
model.fit(train_images, train_labels, epochs=5)
model.save_weights('../models/weight.hdf5')