新农村建设投诉网站,手机建设银行网站首页,现在做网站用的软件,建筑图纸怎么看懂的最快 施工图本文目标
分享李沐老师关于深度学习的观点#xff1a;1⃣️从实践的角度入手深度学习可能比单纯的研究算法更好#xff1b;2⃣️如果想学习深度学习#xff0c;要只用简单的数据结构#xff0c;譬如numpy、NDArray#xff0c;从0实现一个深度学习算法#xff0c;这样才能…本文目标
分享李沐老师关于深度学习的观点1⃣️从实践的角度入手深度学习可能比单纯的研究算法更好2⃣️如果想学习深度学习要只用简单的数据结构譬如numpy、NDArray从0实现一个深度学习算法这样才能碰到进而解决深度学习中的许多核心问题也可以更好的理解现在流行的框架3⃣️从应用的角度那就直接上现成的框架结合真实数据不断练习调得一手好参结合李航《统计学习方法》中的观点总结出机器学习深度学习的一般代码框架具体看代码。
机器学习的一般框架
从0实现版
# -*- coding: utf-8 -*-
import d2lzh as d2l
from mxnet import nd
from mxnet import autograd# data
batch_size 256
train_iter, test_iter d2l.load_data_fashion_mnist(batch_size)# model
num_inputs, num_hiddens1, num_hiddens2, num_outputs 784, 256, 256, 10
W1 nd.random.normal(scale0.01, shape(num_inputs, num_hiddens1))
b1 nd.zeros(num_hiddens1)
W2 nd.random.normal(scale0.01, shape(num_hiddens1, num_hiddens2))
b2 nd.zeros(num_hiddens2)
W3 nd.random.normal(scale0.01, shape(num_hiddens2, num_outputs))
b3 nd.zeros(num_outputs)
params [W1, b1, W2, b2, W3, b3]for param in params:param.attach_grad()def relu(X):return nd.maximum(X, 0)def softmax(X):X_exp X.exp()partition X_exp.sum(axis1, keepdimsTrue)return X_exp / partitiondef net(X):X X.reshape((-1, num_inputs))H1 relu(nd.dot(X, W1) b1)H2 relu(nd.dot(H1, W2) b2)return softmax(H2)# strategy
def cross_entropy(y_hat, y):return -nd.pick(y_hat, y).log()loss cross_entropy# algorithm
def sgd(params, lr, batch_size):for param in params:param[:] param - lr * param.grad / batch_size# training
def evaluate_accuracy(data_iter, net):acc_sum, n 0.0, 0for X, y in data_iter:y y.astype(float32)acc_sum (net(X).argmax(axis1) y).sum().asscalar()n y.sizereturn acc_sum / ndef train(net, train_iter, test_iter, loss, num_epochs, batch_size, params, lr):for epoch in range(num_epochs):train_l_sum, train_acc_sum, n 0.0, 0.0, 0for X, y in train_iter:with autograd.record():y_hat net(X)l loss(y_hat, y).sum()l.backward()sgd(params, lr, batch_size)y y.astype(float32)train_l_sum l.asscalar()train_acc_sum (y_hat.argmax(axis1) y).sum().asscalar()n y.sizetest_acc evaluate_accuracy(test_iter, net)print(epoch: %d, loss %.4f, train_acc %.3f, test_acc %.3f% (epoch 1, train_l_sum / n, train_acc_sum / n, test_acc))num_epochs, lr 10, 0.3
train(net, train_iter, test_iter, loss, num_epochs, batch_size,params, lr)
# predictif __name__ __main__:print(------ok-------)说明代码中还是使用了d2l.load_data_fashion_mnist来加载图片数据有时间把这个也替换掉用NDArray实现
mxnet框架版
# -*- coding: utf- -*-
import d2lzh as d2l
from mxnet import gluon, init
from mxnet.gluon import loss as gloss, nn# data
batch_size 256
train_iter, test_iter d2l.load_data_fashion_mnist(batch_size)# model
net nn.Sequential()
net.add(nn.Dense(256, activationrelu),nn.Dense(256, activationrelu),nn.Dense(10))
net.initialize(init.Normal(sigma0.01))# strategy
loss gloss.SoftmaxCrossEntropyLoss()# algorithm
lr 0.3
trainer gluon.Trainer(net.collect_params(), sgd, {learning_rate: lr})# training
num_epochs 10
d2l.train_ch3(net, train_iter, test_iter, loss, num_epochs, batch_size, None, None, trainer)
# predictif __name__ __main__:print(-----ok------)参考资料
