def test_train_model_svr(data_train_test, model_file_name):
model = make_pipeline(
DropColumn('timeStamp'),
StandardScaler(),
SVR(C=1.e5, gamma=0.005, kernel='rbf'),
)
train_model(data_train_test, model, model_file_name)
def setup_model(self, model):
global status_data
global my_model
status_data['text'] = "Loading data..."
(x_train, y_train), (x_test,
y_test) = tf.keras.datasets.mnist.load_data()
print(type(x_train))
print(x_train.shape)
print(type(x_train[0]))
print(x_train[0].shape)
x_train_normalized = x_train / 255
x_test_normalized = x_test / 255
status_data['text'] = "Training model..."
# The following variables are the hyperparameters.
learning_rate = 0.003
epochs = 50
batch_size = 4000
validation_split = 0.2
# Establish the model's topography.
my_model = model.create_model(learning_rate)
# Train the model on the normalized training set.
epochs, hist = model.train_model(my_model, x_train_normalized, y_train,
epochs, batch_size, validation_split)
# Plot a graph of the metric vs. epochs.
list_of_metrics_to_plot = ['accuracy']
model.plot_curve(epochs, hist, list_of_metrics_to_plot)
# Evaluate against the test set.
print("\n Evaluate the new model against the test set:")
my_model.evaluate(x=x_test_normalized, y=y_test, batch_size=batch_size)
status_data['text'] = "Model trained."
dev_images1 = dev_images1.astype('float32') / 255
# general categoricla variable
# notice I am add "s" at train_label, to distinguish from earlier numpy.
train_labels = to_categorical(train_label_np)
dev_labels = to_categorical(dev_label_np)
train_labels1 = to_categorical(train_label_np1)
dev_labels1 = to_categorical(dev_label_np1)
train_labels_merged = merge_labels(train_labels, train_labels1)
dev_labels_merged = merge_labels(dev_labels, dev_labels1)
sample_size = dev_labels_merged.shape[0] # sample size
for i in range(sample_size):
print(train_labels_merged[i, :], ",", train_label_np[i], ",",
train_label_np1[i])
params = {}
params['height'] = height
params['width'] = width
params['channel'] = channel
model = build_model(True, params)
history = train_model(model, train_labels_merged, train_images,
dev_labels_merged, dev_images)
# print the graph of learning history for diagnostic purpose.
print_plot_keras_metrics(history)
x2 = data.normalize(x2, 0)
n_clusters = len(set(gt))
#act_ae1, act_ae2, act_dg1, act_dg2 = 'sigmoid', 'sigmoid', 'sigmoid', 'sigmoid'
v1_aedims_ = [[x1.shape[1], 512, 256], [256, 512, x1.shape[1]]]
v2_aedims_ = [[x2.shape[1], 512, 256], [256, 512, x2.shape[1]]]
#原来的
v1_aedims_ = [[x1.shape[1], 512, 128], [128, 512, x1.shape[1]]]
v2_aedims_ = [[x2.shape[1], 512, 128], [128, 512, x2.shape[1]]]
mae_dims_ = [[256, 128, 64], [256, 128, 64], [64, 128, 256],
[64, 128, 256]]
mae_dims_ = [[128, 128, 64], [128, 128, 64], [64, 128, 128],
[64, 128, 128]]
#现在用的
#dims_dg1 = [64, 100]
#dims_dg2 = [64, 100]
dis_dims_ = [256, 128, 1]
para_lambda = 1
batch_size = 256
epochs = 200
model = MaeAEModel(v1_aedims=v1_aedims_,
v2_aedims=v2_aedims_,
mae_dims=mae_dims_,
dis_dims=dis_dims_) #duaAE用的
H, gt = model.train_model(x1, x2, gt, epochs, batch_size)
#H,gt=model(x1, x2, gt, para_lambda, dims, act, lr, epochs, batch_size)
print_result(n_clusters, H, gt)
path="./output/model_deep.h5",
mask=False,
)
model.cnn_rnn_attn(hiddenDim=400,
)
train_x = [train_input,train_char,]
train_y = train_label
test_x = [test_input,test_char,]
# train
print("train...")
epoch = 5
model.train_model(x=train_x,y=train_y,
epoch=epoch,
batch_size=16,
validation_split=0.1,
)
print("predict...")
test_predict = model.predict(test_x)
print(type(test_predict))
print(test_predict.shape)
test_predict = np.argmax(test_predict, axis=2)
pred_tag = []
for i, tokens in enumerate(test_token):
tag_sent = []
for j, t in enumerate(tokens):
try:
tag_sent.append(tag_alphabet.get_instance(test_predict[i][j]))
except IndexError:
continue
def test_train_model_mlp(data_train_test, model_file_name):
model = make_pipeline(DropColumn('timeStamp'), StandardScaler(),
MLPRegressor(alpha=1.e-4, learning_rate_init=10.))
train_model(data_train_test, model, model_file_name)
