def exp_dense():
np.random.seed(10)
cuda.set_cuda_active(False)
a = np.random.rand(32, 320).astype(np.float32)
b = np.random.rand(32, 80).astype(np.float32)
layer1 = rm.Dense(input_size=320, output_size=100)
layer2 = rm.Dense(input_size=100, output_size=80)
ga = rm.Variable(a, auto_update=False)
gb = rm.Variable(b, auto_update=False)
opt = Sgd(0.01, momentum=0.3)
start_t = time.time()
for _ in range(500):
loss = rm.Sum((layer2(rm.Sigmoid(layer1(ga))) - gb)**2) / 32
loss.ensure_cpu()
print(loss)
grad = loss.grad()
grad.update(opt)
print(time.time() - start_t)
def exp_convolution2():
np.random.seed(10)
cuda.set_cuda_active(True)
a = np.random.randn(8, 3, 12, 12).astype(np.float32)
b = np.random.randn(8, 16, 10, 10).astype(np.float32)
layer1 = rm.Conv2d(channel=16, input_size=a.shape[1:])
ga = rm.Variable(a, auto_update=False)
gb = rm.Variable(b, auto_update=False)
opt = Sgd(0.001, momentum=0.3)
start_t = time.time()
for _ in range(100000):
loss = rm.Sum((rm.Sigmoid(layer1(ga)) - gb)**2) / 8
loss.ensure_cpu()
print(loss)
grad = loss.grad()
grad.update(opt)
del loss
print(time.time() - start_t)
def exp_convolution1():
np.random.seed(10)
# Caused by CUDNN_CONVOLUTION_FWD_ALGO_GEMM is not deterministic.
# 1724.07080078 GPU
# 1715.86767578 CPU
cuda.set_cuda_active(True)
a = np.random.randn(8 * 2, 64, 32, 32).astype(np.float32)
b = np.random.randn(8 * 2, 32, 28, 28).astype(np.float32)
layer1 = rm.Conv2d(channel=32, input_size=a.shape[1:])
layer2 = rm.Conv2d(channel=32, input_size=(32, 30, 30))
ga = rm.Variable(a, auto_update=False)
gb = rm.Variable(b, auto_update=False)
opt = Sgd(0.0001, momentum=0.0)
start_t = time.time()
for _ in range(100):
loss = rm.Sum((layer2(rm.Relu(layer1(ga))) - gb)**2) / 8
loss.ensure_cpu()
grad = loss.grad()
grad.update(opt)
print(loss)
print(time.time() - start_t)
print("")
t3 = self.layer2(t2)
print("hidden x output weight:\n{}".format(self.layer2.params.w))
print("hidden x output bias:\n{}".format(self.layer2.params.b))
print("")
print("output:\n{}".format(t3))
print("output shape:{}".format(t3.shape))
print("")
return t3
epoch = 50
batch = 1
N = len(X)
optimizer = Sgd()
network = Mnist()
learning_curve = []
for i in range(epoch):
perm = np.random.permutation(N)
loss = 0
for j in range(0, N // batch):
train_batch = X[perm[j * batch:(j + 1) * batch]]
response_batch = y[perm[j * batch:(j + 1) * batch]]
with network.train():
result = network(train_batch)
l = rm.sigmoid_cross_entropy(result, response_batch)
grad = l.grad()
grad.update(optimizer)
def __init__(self, epoch, batch_size, network, opt=Sgd(), verbose=0):
self.epoch = epoch
self.batch_size = batch_size
self.network = network
self.opt = opt
self.verbose = verbose
class Mnist(rm.Model):
def __init__(self):
self.layer1 = rm.Dense(output_size=5)
self.layer2 = rm.Dense(1)
def forward(self, x):
t1 = self.layer1(x)
t2 = rm.relu(t1)
t3 = self.layer2(t2)
return t3
epoch = 50
batch = 1
N = len(X)
optimizer = Sgd(lr=0.1, momentum=0.4)
network = Mnist()
learning_curve = []
for i in range(epoch):
perm = np.random.permutation(N)
loss = 0
for j in range(0, N // batch):
train_batch = X[perm[j * batch:(j + 1) * batch]]
response_batch = y[perm[j * batch:(j + 1) * batch]]
with network.train():
result = network(train_batch)
l = rm.sigmoid_cross_entropy(result, response_batch)
grad = l.grad()
grad.update(optimizer)
self.layer1 = rm.Dense(output_size=100)
self.layer2 = rm.Dense(output_size=10)
def forward(self, x):
return self.layer2(rm.relu(self.layer1(x)))
epoch = 10
batch = 100
count = 0
learning_curve = []
test_learning_curve = []
opt = Adam()
opt = Sgd()
N = len(X_train)
nn = MNist()
for i in range(epoch):
start_t = time.time()
perm = np.random.permutation(N)
loss = 0
for j in range(0, N // batch):
train_batch = X_train[perm[j * batch:(j + 1) * batch]]
responce_batch = labels_train[perm[j * batch:(j + 1) * batch]]
with nn.train():
result = nn(train_batch)
classes = 10
bbox = 5
# load image generator
print("loading image generator...")
generator = ImageGenerator(item_path, background_path)
# load model
print("loading initial model...")
model = YOLOv2(classes=classes, bbox=bbox)
model.load(initial_weight_file)
num_gpu = cuGetDeviceCount()
#model.to_gpu()
opt = Sgd(lr=learning_rate, momentum=momentum)
#trainer = Trainer(model, batch_size=32, loss_func=yolo_detector, num_epoch=1, optimizer=opt, num_gpu=num_gpu)
#input_width=input_height=320
#x, t = generator.generate_samples(
# n_samples=16,
# n_items=3,
# crop_width=input_width,
# crop_height=input_height,
# min_item_scale=0.1,
# max_item_scale=0.4,
# rand_angle=25,
# minimum_crop=0.8,
# delta_hue=0.01,
# delta_sat_scale=0.5,
print("")
t3 = self.layer2(t2)
print("hidden x output weight:\n{}".format(self.layer2.params.w))
print("hidden x output bias:\n{}".format(self.layer2.params.b))
print("")
print("output:\n{}".format(t3))
print("output shape:{}".format(t3.shape))
print("")
return t3
epoch = 1000
batch = 1
N = len(X)
optimizer = Sgd(lr=0.05, momentum=0.0)
network = Mnist()
learning_curve = []
for i in range(epoch):
perm = np.random.permutation(N)
loss = 0
for j in range(0, N // batch):
train_batch = X[perm[j * batch:(j + 1) * batch]]
response_batch = y[perm[j * batch:(j + 1) * batch]]
with network.train():
result = network(train_batch)
#l = rm.sigmoid_cross_entropy(result, response_batch)
l = rm.mse(result, response_batch)
grad = l.grad()