# train_set = my_framework.datasets.MNIST(train=True, transform=f)
# test_set = my_framework.datasets.MNIST(train=False, transform=f)
max_epoch = 200
batch_size = 100
hidden_size = 1000
bit_size = 1
train_set = my_framework.datasets.MNIST(train=True)
test_set = my_framework.datasets.MNIST(train=False)
train_loader = DataLoader(train_set, batch_size)
test_loader = DataLoader(test_set, batch_size, shuffle=False)
model = MLP((hidden_size, 10))
optimizer = optimizers.SGD().setup(model)
train_acc = np.zeros(max_epoch)
test_acc = np.zeros(max_epoch)
train_loss = np.zeros(max_epoch)
test_loss = np.zeros(max_epoch)
for epoch in range(max_epoch):
sum_loss, sum_acc = 0, 0
for x, t in train_loader:
y = model(x)
loss = F.softmax_cross_entropy_simple(y, t)
acc = F.accuracy(y, t)
model.cleargrads()
loss.backward()
sys.path.append(os.path.join(os.path.dirname(__file__), '..'))
import numpy as np
from my_framework import Variable, Model, optimizers
import my_framework.functions as F
import my_framework.layers as L
from my_framework.models import MLP
np.random.seed(0)
x = np.random.rand(100, 1)
y = np.sin(2 * np.pi * x) + np.random.rand(100, 1)
lr = 0.2
iters = 10000
hidden_size = 10
model = MLP((hidden_size, 1))
optimizer = optimizers.SGD(lr)
optimizer.setup(model)
for i in range(iters):
y_pred = model(x)
loss = F.mean_squared_error(y, y_pred)
model.cleargrads()
loss.backward()
optimizer.update()
if i % 1000 == 0:
print(loss)