Pytorch torch.optim优化器个性化使用

一、简化前馈网络LeNet

二、优化器基本使用方法

1. 建立优化器实例
2. 循环：
   1. 清空梯度
   2. 向前传播
   3. 计算Loss
   4. 反向传播
   5. 更新参数

三、网络模块参数定制

为不同的子网络参数不同的学习率，finetune常用，使分类器学习率参数更高，学习速度更快（理论上）。

1.经由构建网络时划分好的模组进行学习率设定，

2.以网络层对象为单位进行分组，并设定学习率

四、在训练中动态的调整学习率

可以看到optimizer.param_groups结构，[{'params','lr', 'momentum', 'dampening', 'weight_decay', 'nesterov'},{……}]，集合了优化器的各项参数。

• torch.optim的灵活使用

• 重写sgd优化器

import torch
from torch.optim.optimizer import Optimizer, required

class SGD(Optimizer):
    def __init__(self, params, lr=required, momentum=0, dampening=0, weight_decay1=0, weight_decay2=0, nesterov=False):
        defaults = dict(lr=lr, momentum=momentum, dampening=dampening,
                        weight_decay1=weight_decay1, weight_decay2=weight_decay2, nesterov=nesterov)
        if nesterov and (momentum <= 0 or dampening != 0):
            raise ValueError("Nesterov momentum requires a momentum and zero dampening")
        super(SGD, self).__init__(params, defaults)

    def __setstate__(self, state):
        super(SGD, self).__setstate__(state)
        for group in self.param_groups:
            group.setdefault('nesterov', False)

    def step(self, closure=None):
        """Performs a single optimization step. Arguments: closure (callable, optional): A closure that reevaluates the model and returns the loss. """
        loss = None
        if closure is not None:
            loss = closure()

        for group in self.param_groups:
            weight_decay1 = group['weight_decay1']
            weight_decay2 = group['weight_decay2']
            momentum = group['momentum']
            dampening = group['dampening']
            nesterov = group['nesterov']

            for p in group['params']:
                if p.grad is None:
                    continue
                d_p = p.grad.data
                if weight_decay1 != 0:
                    d_p.add_(weight_decay1, torch.sign(p.data))
                if weight_decay2 != 0:
                    d_p.add_(weight_decay2, p.data)
                if momentum != 0:
                    param_state = self.state[p]
                    if 'momentum_buffer' not in param_state:
                        buf = param_state['momentum_buffer'] = torch.zeros_like(p.data)
                        buf.mul_(momentum).add_(d_p)
                    else:
                        buf = param_state['momentum_buffer']
                        buf.mul_(momentum).add_(1 - dampening, d_p)
                    if nesterov:
                        d_p = d_p.add(momentum, buf)
                    else:
                        d_p = buf

                p.data.add_(-group['lr'], d_p)

        return loss