学习率调整

学习率(learning rate)控制更新的步伐

学习率调整

class_LRScheduler
主要属性
optimizer 关联的优化器
last_epoch 记录epoch数
base_lrs 记录初始学习率

主要方法

step() 更新下一个epoch的学习率
get_lr() 虚函数，计算下一个epoch的学习率

1.StepLR

功能等间隔调整学习率
主要参数
step_size 调整间隔数
gamma 调整系数
调整方式
lr = lr*gamma

2.MultiStepLR

功能按给定间隔调整学习率
主要参数
milestones 设定调整时刻数
gamma 调整系数
调整方式
lr = lr*gamma

3.ExponentialLR

功能按指数衰减调整学习率
主要参数
gamma 指数的底
调整方式 lr = lr*gamma ** epoch

4.CosineAnnealingLR

功能余弦周期调整学习率
主要参数
T_max 下降周期
eta_min 学习率下限
调整方式
$\eta_{t}=\eta_{\min }+\frac{1}{2}\left(\eta_{\max }-\eta_{\min }\right)\left(1+\cos \left(\frac{T_{\operatorname{cur}}}{T_{\max }} \pi\right)\right)$

5.ReduceLRonPlateau

功能监控指标，当指标不再变化则调整
主要参数：
mode min/max 两种模式
factor 调整系数
patience “耐心”，接受几次不变化
cooldown “冷却时间”，停止监控一段时间
verbose 是否打印日志
min_lr 学习率下限
eps 学习率衰减最小值

6.LambdaLR

功能自定义调整策略
主要参数
lr_lambda function or list

学习率调整小结

1.有序调整 Step,MultiStep,Exponential,CosineAnnealing
2.自适应调整 ReduceLROnPleateau
3.自定义调整 Lambda

学习率初始化

1.设置较小数：0.01、0.001、0.0001
2.搜索最大学习率:《Cyclical Learning Rates for Training Neural Networks》

import torch
import torch.optim as optim
import numpy as np
import matplotlib.pyplot as plt
torch.manual_seed(1)

LR = 0.1
iteration = 10
max_epoch = 200
# ------------------------------ fake data and optimizer  ------------------------------

weights = torch.randn((1), requires_grad=True)
target = torch.zeros((1))

optimizer = optim.SGD([weights], lr=LR, momentum=0.9)

# ------------------------------ 1 Step LR ------------------------------
# flag = 0
flag = 1
if flag:

    scheduler_lr = optim.lr_scheduler.StepLR(optimizer, step_size=50, gamma=0.1)  # 设置学习率下降策略

    lr_list, epoch_list = list(), list()
    for epoch in range(max_epoch):

        lr_list.append(scheduler_lr.get_lr())
        epoch_list.append(epoch)

        for i in range(iteration):

            loss = torch.pow((weights - target), 2)
            loss.backward()

            optimizer.step()
            optimizer.zero_grad()

        scheduler_lr.step()

    plt.plot(epoch_list, lr_list, label="Step LR Scheduler")
    plt.xlabel("Epoch")
    plt.ylabel("Learning rate")
    plt.legend()
    plt.show()


# ------------------------------ 2 Multi Step LR ------------------------------
flag = 0
# flag = 1
if flag:

    milestones = [50, 125, 160]
    scheduler_lr = optim.lr_scheduler.MultiStepLR(optimizer, milestones=milestones, gamma=0.1)

    lr_list, epoch_list = list(), list()
    for epoch in range(max_epoch):

        lr_list.append(scheduler_lr.get_lr())
        epoch_list.append(epoch)

        for i in range(iteration):

            loss = torch.pow((weights - target), 2)
            loss.backward()

            optimizer.step()
            optimizer.zero_grad()

        scheduler_lr.step()

    plt.plot(epoch_list, lr_list, label="Multi Step LR Scheduler\nmilestones:{}".format(milestones))
    plt.xlabel("Epoch")
    plt.ylabel("Learning rate")
    plt.legend()
    plt.show()


# ------------------------------ 3 Exponential LR ------------------------------
flag = 0
# flag = 1
if flag:

    gamma = 0.95
    scheduler_lr = optim.lr_scheduler.ExponentialLR(optimizer, gamma=gamma)

    lr_list, epoch_list = list(), list()
    for epoch in range(max_epoch):

        lr_list.append(scheduler_lr.get_lr())
        epoch_list.append(epoch)

        for i in range(iteration):

            loss = torch.pow((weights - target), 2)
            loss.backward()

            optimizer.step()
            optimizer.zero_grad()

        scheduler_lr.step()

    plt.plot(epoch_list, lr_list, label="Exponential LR Scheduler\ngamma:{}".format(gamma))
    plt.xlabel("Epoch")
    plt.ylabel("Learning rate")
    plt.legend()
    plt.show()

# ------------------------------ 4 Cosine Annealing LR ------------------------------
flag = 0
# flag = 1
if flag:

    t_max = 50
    scheduler_lr = optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=t_max, eta_min=0.)

    lr_list, epoch_list = list(), list()
    for epoch in range(max_epoch):

        lr_list.append(scheduler_lr.get_lr())
        epoch_list.append(epoch)

        for i in range(iteration):

            loss = torch.pow((weights - target), 2)
            loss.backward()

            optimizer.step()
            optimizer.zero_grad()

        scheduler_lr.step()

    plt.plot(epoch_list, lr_list, label="CosineAnnealingLR Scheduler\nT_max:{}".format(t_max))
    plt.xlabel("Epoch")
    plt.ylabel("Learning rate")
    plt.legend()
    plt.show()


# ------------------------------ 5 Reduce LR On Plateau ------------------------------
# flag = 0
flag = 1
if flag:
    loss_value = 0.5
    accuray = 0.9

    factor = 0.1
    mode = "min"
    patience = 10  # 若10个epoch都不下降 就更新
    cooldown = 10
    min_lr = 1e-4
    verbose = True

    scheduler_lr = optim.lr_scheduler.ReduceLROnPlateau(optimizer, factor=factor, mode=mode, patience=patience,
                                                        cooldown=cooldown, min_lr=min_lr, verbose=verbose)

    for epoch in range(max_epoch):
        for i in range(iteration):

            # train(...)

            optimizer.step()
            optimizer.zero_grad()

        if epoch == 5:
            loss_value = 0.4

        scheduler_lr.step(loss_value)


# ------------------------------ 6 lambda ------------------------------
# flag = 0
flag = 1
if flag:

    lr_init = 0.1

    weights_1 = torch.randn((6, 3, 5, 5))
    weights_2 = torch.ones((5, 5))

    optimizer = optim.SGD([
        {'params': [weights_1]},
        {'params': [weights_2]}], lr=lr_init)

    lambda1 = lambda epoch: 0.1 ** (epoch // 20)
    lambda2 = lambda epoch: 0.95 ** epoch

    scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer, lr_lambda=[lambda1, lambda2])

    lr_list, epoch_list = list(), list()
    for epoch in range(max_epoch):
        for i in range(iteration):

            # train(...)

            optimizer.step()
            optimizer.zero_grad()

        scheduler.step()

        lr_list.append(scheduler.get_lr())
        epoch_list.append(epoch)

        print('epoch:{:5d}, lr:{}'.format(epoch, scheduler.get_lr()))

    plt.plot(epoch_list, [i[0] for i in lr_list], label="lambda 1")
    plt.plot(epoch_list, [i[1] for i in lr_list], label="lambda 2")
    plt.xlabel("Epoch")
    plt.ylabel("Learning Rate")
    plt.title("LambdaLR")
    plt.legend()
    plt.show()

来源：CSDN

作者：TongYixuan_LUT

链接：https://blog.csdn.net/qq_33357094/article/details/104576810

标签

易学教程内所有资源均来自网络或用户发布的内容，如有违反法律规定的内容欢迎反馈！
该文章没有解决你所遇到的问题?点击提问,说说你的问题,让更多的人一起探讨吧!