def main():
(x_train, t_train), (x_test, t_test) = load_mnist(flatten=False)
print(x_train.shape, t_train.shape)
print(t_train[0])
net = MultiLayerNet(is_use_dropout=False)
net.add_layer(Layer.Conv2D(16, (3, 3), pad=1, input_size=(1, 28, 28)), initializer=Initializer.He(),
activation=Layer.Relu())
net.add_layer(Layer.BatchNormalization())
net.add_layer(Layer.Relu())
net.add_layer(Layer.Pooling(pool_h=2, pool_w=2, stride=2))
net.add_layer(Layer.Conv2D(16, (3, 3), pad=1, initializer=Initializer.He()))
net.add_layer(Layer.BatchNormalization())
net.add_layer(Layer.Relu())
net.add_layer(Layer.Pooling(pool_h=2, pool_w=2, stride=2))
net.add_layer(Layer.Dense(20, initializer=Initializer.He(), activation=Layer.Relu()))
net.add_layer(Layer.Dropout(0.5))
net.add_layer(Layer.Dense(10))
net.add_layer(Layer.Dropout(0.5))
net.add_layer(Layer.SoftmaxWithLoss())
if gpu_enable:
net.to_gpu()
for k, v in net.params.items():
print(k, v.shape)
result = net.train(
x_train, t_train, x_test, t_test, batch_size=200, iters_num=100, print_epoch=1, evaluate_limit=500,
is_use_progress_bar=True,
optimizer=Optimizer.Adam(lr=0.001))
import pickle
import datetime
## Save pickle
with open(f"train_data_{str(datetime.datetime.now())[:-7].replace(':', '')}.pickle", "wb") as fw:
pickle.dump(result, fw)
# net.save_model()
print("============================================")
net = MultiLayerNet()
net.add_layer(Layer.Dense(1, input_size=2,
activation=Layer.IdentityWithLoss()))
# net.add_layer(Layer.Dense(5, input_size = 2, activation=Layer.Relu() ))
# net.add_layer(Layer.Dense(1))
x_train, t_train, x_test, t_test = shuffle_split_data(x_data, t_data, 0.2)
print(net.params)
# scaler = Scaler.StandardScaler()
# x_train = scaler.fit_transform(x_train)
# x_test = scaler.transform(x_test)
result = net.train(x_train,
t_train,
x_test,
t_test,
batch_size=100,
iters_num=1000,
print_epoch=30,
optimizer=Optimizer.SGD(lr=0.001))
print("done!")
print(net.params)
print(net.layers["Affine0"].__dict__)
net.save_model()
# net.load_model("weight.npz")
net.add_layer(Layer.BatchNormalization())
net.add_layer(Layer.Relu())
net.add_layer(Layer.Dense(64), initializer=Initializer.He())
net.add_layer(Layer.BatchNormalization())
net.add_layer(Layer.Relu())
net.add_layer(
Layer.Dense(10,
initializer=Initializer.He(),
activation=Layer.SoftmaxWithLoss()))
result = net.train(x_train,
t_train,
x_test,
t_test,
batch_size=300,
iters_num=1000,
print_epoch=1,
is_use_progress_bar=True,
save_model_each_epoch=1,
save_model_path="./dog_cat_result",
optimizer=Optimizer.Adam(lr=0.01))
print("done!")
print(net.params)
# print(net.layers)
# net.save_model()
# net.load_model("weight.npz")