多层感知机——gluon

多层感知机

多层感知机与前面介绍的多类逻辑回归非常类似，主要的区别是我们在输入层和输出层之间插入了一个到多个隐含层。

定义一个只有一个隐含层的模型，这个隐含层输出256个节点。

import sys
sys.path.append('..')
import utils
batch_size = 256
train_data, test_data = utils.load_data_fashion_mnist(batch_size)

D:\Users\Administrator\Anaconda3\lib\site-packages\h5py\__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
  from ._conv import register_converters as _register_converters
D:\Users\Administrator\Anaconda3\lib\site-packages\mxnet\gluon\data\vision\datasets.py:84: DeprecationWarning: The binary mode of fromstring is deprecated, as it behaves surprisingly on unicode inputs. Use frombuffer instead
  label = np.fromstring(fin.read(), dtype=np.uint8).astype(np.int32)
D:\Users\Administrator\Anaconda3\lib\site-packages\mxnet\gluon\data\vision\datasets.py:88: DeprecationWarning: The binary mode of fromstring is deprecated, as it behaves surprisingly on unicode inputs. Use frombuffer instead
  data = np.fromstring(fin.read(), dtype=np.uint8)

In [2]:

from mxnet import ndarray as nd

num_inputs = 28*28
num_outputs = 10

num_hidden = 256
weight_scale = .01

W1 = nd.random_normal(shape=(num_inputs, num_hidden), scale=weight_scale)
b1 = nd.zeros(num_hidden)

W2 = nd.random_normal(shape=(num_hidden, num_outputs), scale=weight_scale)
b2 = nd.zeros(num_outputs)

params = [W1, b1, W2, b2]

for param in params:
    param.attach_grad()

In [3]:

def relu(X):
    return nd.maximum(X, 0)
def net(X):
    X = X.reshape((-1, num_inputs))
    h1 = relu(nd.dot(X, W1) + b1)
    output = nd.dot(h1, W2) + b2
    return output

from mxnet import gluon
softmax_cross_entropy = gluon.loss.SoftmaxCrossEntropyLoss()

In [4]:

from mxnet import autograd as autograd

learning_rate = .5

for epoch in range(5):
    train_loss = 0.
    train_acc = 0.
    for data, label in train_data:
        with autograd.record():
            output = net(data)
            loss = softmax_cross_entropy(output, label)
        loss.backward()
        utils.SGD(params, learning_rate/batch_size)

        train_loss += nd.mean(loss).asscalar()
        train_acc += utils.accuracy(output, label)

    test_acc = utils.evaluate_accuracy(test_data, net)
    print("Epoch %d. Loss: %f, Train acc %f, Test acc %f" % (
        epoch, train_loss/len(train_data),
        train_acc/len(train_data), test_acc))

Epoch 0. Loss: 0.817416, Train acc 0.695663, Test acc 0.750601
Epoch 1. Loss: 0.489647, Train acc 0.818743, Test acc 0.844952
Epoch 2. Loss: 0.433992, Train acc 0.839744, Test acc 0.856871
Epoch 3. Loss: 0.400036, Train acc 0.851846, Test acc 0.865585
Epoch 4. Loss: 0.375272, Train acc 0.861545, Test acc 0.846855

********************************************************************************************************************

********************************************************************************************************************

使用gluon的多层感知机（MLP）

In [5]:

#使用gluon
from mxnet import gluon

net = gluon.nn.Sequential()
with net.name_scope():
    net.add(gluon.nn.Dense(256, activation="relu"))
    net.add(gluon.nn.Dense(10))
net.initialize()

In [6]:

import sys
sys.path.append('..')
from mxnet import ndarray as nd
from mxnet import autograd
import utils


batch_size = 256
train_data, test_data = utils.load_data_fashion_mnist(batch_size)

softmax_cross_entropy = gluon.loss.SoftmaxCrossEntropyLoss()
trainer = gluon.Trainer(net.collect_params(), 'sgd', {'learning_rate': 0.5})

for epoch in range(5):
    train_loss = 0.
    train_acc = 0.
    for data, label in train_data:
        with autograd.record():
            output = net(data)
            loss = softmax_cross_entropy(output, label)
        loss.backward()
        trainer.step(batch_size)

        train_loss += nd.mean(loss).asscalar()
        train_acc += utils.accuracy(output, label)

    test_acc = utils.evaluate_accuracy(test_data, net)
    print("Epoch %d. Loss: %f, Train acc %f, Test acc %f" % (
        epoch, train_loss/len(train_data), train_acc/len(train_data), test_acc))

D:\Users\Administrator\Anaconda3\lib\site-packages\mxnet\gluon\data\vision\datasets.py:84: DeprecationWarning: The binary mode of fromstring is deprecated, as it behaves surprisingly on unicode inputs. Use frombuffer instead
  label = np.fromstring(fin.read(), dtype=np.uint8).astype(np.int32)
D:\Users\Administrator\Anaconda3\lib\site-packages\mxnet\gluon\data\vision\datasets.py:88: DeprecationWarning: The binary mode of fromstring is deprecated, as it behaves surprisingly on unicode inputs. Use frombuffer instead
  data = np.fromstring(fin.read(), dtype=np.uint8)

Epoch 0. Loss: 0.716115, Train acc 0.735276, Test acc 0.773237
Epoch 1. Loss: 0.468759, Train acc 0.827424, Test acc 0.853466
Epoch 2. Loss: 0.421417, Train acc 0.845553, Test acc 0.798578
Epoch 3. Loss: 0.389830, Train acc 0.855669, Test acc 0.870593
Epoch 4. Loss: 0.361950, Train acc 0.865001, Test acc 0.857272