resent_torch.py

# -*- coding: utf-8 -*-
"""
# Author  : Ming
# File    : {NAME}.py
# Time    : 2019/5/10 0010 下午 4:22
"""

import torch
import torch.nn as nn
from resnets_utils import *
import numpy as np
import torch
import torch.nn as nn
import torch.optim as optim
import torchvision
import torchvision.transforms as transforms
from torch.optim.lr_scheduler import _LRScheduler
import os
from torch.utils.data import Dataset
from torch.utils.data import DataLoader
from torch.autograd import Variable


class BasicBlock(nn.Module):
    """Basic Block for resnet 18 and resnet 34
    """

    # BasicBlock and BottleNeck block
    # have different output size
    # we use class attribute expansion
    # to distinct
    expansion = 1

    def __init__(self, in_channels, out_channels, stride=1):
        super().__init__()

        # residual function
        self.residual_function = nn.Sequential(
            nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=stride, padding=1, bias=False),
            nn.BatchNorm2d(out_channels),
            nn.ReLU(inplace=True),
            nn.Conv2d(out_channels, out_channels * BasicBlock.expansion, kernel_size=3, padding=1, bias=False),
            nn.BatchNorm2d(out_channels * BasicBlock.expansion)
        )

        # shortcut
        self.shortcut = nn.Sequential()

        # the shortcut output dimension is not the same with residual function
        # use 1*1 convolution to match the dimension
        if stride != 1 or in_channels != BasicBlock.expansion * out_channels:
            self.shortcut = nn.Sequential(
                nn.Conv2d(in_channels, out_channels * BasicBlock.expansion, kernel_size=1, stride=stride, bias=False),
                nn.BatchNorm2d(out_channels * BasicBlock.expansion)
            )

    def forward(self, x):
        return nn.ReLU(inplace=True)(self.residual_function(x) + self.shortcut(x))


class BottleNeck(nn.Module):
    """Residual block for resnet over 50 layers
    """
    expansion = 4

    def __init__(self, in_channels, out_channels, stride=1):
        super().__init__()
        self.residual_function = nn.Sequential(
            nn.Conv2d(in_channels, out_channels, kernel_size=1, bias=False),
            nn.BatchNorm2d(out_channels),
            nn.ReLU(inplace=True),
            nn.Conv2d(out_channels, out_channels, stride=stride, kernel_size=3, padding=1, bias=False),
            nn.BatchNorm2d(out_channels),
            nn.ReLU(inplace=True),
            nn.Conv2d(out_channels, out_channels * BottleNeck.expansion, kernel_size=1, bias=False),
            nn.BatchNorm2d(out_channels * BottleNeck.expansion),
        )

        self.shortcut = nn.Sequential()

        if stride != 1 or in_channels != out_channels * BottleNeck.expansion:
            self.shortcut = nn.Sequential(
                nn.Conv2d(in_channels, out_channels * BottleNeck.expansion, stride=stride, kernel_size=1, bias=False),
                nn.BatchNorm2d(out_channels * BottleNeck.expansion)
            )

    def forward(self, x):
        return nn.ReLU(inplace=True)(self.residual_function(x) + self.shortcut(x))


class ResNet(nn.Module):

    def __init__(self, block, num_block, num_classes=6):
        super().__init__()

        self.in_channels = 64

        self.conv1 = nn.Sequential(
            nn.Conv2d(3, 64, kernel_size=3, padding=1, bias=False),
            nn.BatchNorm2d(64),
            nn.ReLU(inplace=True))
        # we use a different inputsize than the original paper
        # so conv2_x's stride is 1
        self.conv2_x = self._make_layer(block, 64, num_block[0], 1)
        self.conv3_x = self._make_layer(block, 128, num_block[1], 2)
        self.conv4_x = self._make_layer(block, 256, num_block[2], 2)
        self.conv5_x = self._make_layer(block, 512, num_block[3], 2)
        self.avg_pool = nn.AdaptiveAvgPool2d((1, 1))
        self.fc = nn.Linear(512 * block.expansion, num_classes)

    def _make_layer(self, block, out_channels, num_blocks, stride):
        """make resnet layers(by layer i didnt mean this 'layer' was the
        same as a neuron netowork layer, ex. conv layer), one layer may
        contain more than one residual block
        Args:
            block: block type, basic block or bottle neck block
            out_channels: output depth channel number of this layer
            num_blocks: how many blocks per layer
            stride: the stride of the first block of this layer

        Return:
            return a resnet layer
        """

        # we have num_block blocks per layer, the first block
        # could be 1 or 2, other blocks would always be 1
        strides = [stride] + [1] * (num_blocks - 1)
        layers = []
        for stride in strides:
            layers.append(block(self.in_channels, out_channels, stride))
            self.in_channels = out_channels * block.expansion

        return nn.Sequential(*layers)

    def forward(self, x):
        output = self.conv1(x)
        output = self.conv2_x(output)
        output = self.conv3_x(output)
        output = self.conv4_x(output)
        output = self.conv5_x(output)
        output = self.avg_pool(output)
        output = output.view(output.size(0), -1)
        output = self.fc(output)

        return output


def resnet18():
    """ return a ResNet 18 object
    """
    return ResNet(BasicBlock, [2, 2, 2, 2])


def resnet34():
    """ return a ResNet 34 object
    """
    return ResNet(BasicBlock, [3, 4, 6, 3])


def resnet50():
    """ return a ResNet 50 object
    """
    return ResNet(BottleNeck, [3, 4, 6, 3])


def resnet101():
    """ return a ResNet 101 object
    """
    return ResNet(BottleNeck, [3, 4, 23, 3])


def resnet152():
    """ return a ResNet 152 object
    """
    return ResNet(BottleNeck, [3, 8, 36, 3])


class WarmUpLR(_LRScheduler):
    """warmup_training learning rate scheduler
    Args:
        optimizer: optimzier(e.g. SGD)
        total_iters: totoal_iters of warmup phase
    """

    def __init__(self, optimizer, total_iters, last_epoch=-1):
        self.total_iters = total_iters
        super().__init__(optimizer, last_epoch)

    def get_lr(self):
        """we will use the first m batches, and set the learning
        rate to base_lr * m / total_iters
        """
        return [base_lr * self.last_epoch / (self.total_iters + 1e-8) for base_lr in self.base_lrs]


class PointDataSet(Dataset):
    def __init__(self, x, y):
        self.x_data = x
        self.y_data = y
        self.lenth = x.shape[0]
    def __getitem__(self, index):
        return self.x_data[index], self.y_data[index]
    def __len__(self):
        return self.lenth


if __name__ == '__main__':

    X_train_orig, Y_train_orig, X_test_orig, Y_test_orig, classes = load_dataset()

    # Normalize image vectors
    X_train = X_train_orig / 255.
    X_test = X_test_orig / 255.

    # Convert training and test labels to one hot matrices
    # Y_train = convert_to_one_hot(Y_train_orig, 6).T
    # Y_test = convert_to_one_hot(Y_test_orig, 6).T
    Y_train = Y_train_orig.T
    Y_test = Y_test_orig.T

    print("number of training examples = " + str(X_train.shape[0]))
    print("number of test examples = " + str(X_test.shape[0]))
    print("X_train shape: " + str(X_train.shape))
    print("Y_train shape: " + str(Y_train.shape))
    print("X_test shape: " + str(X_test.shape))
    print("Y_test shape: " + str(Y_test.shape))

    net = resnet18()
    loss_function = nn.CrossEntropyLoss()
    optimizer = optim.SGD(net.parameters(), lr=1e-3, momentum=0.9, weight_decay=5e-4)
    train_scheduler = optim.lr_scheduler.MultiStepLR(optimizer, milestones=[60, 120, 160], gamma=0.2)  # learning rate decay
    batches = 10
    iter_per_epoch = 10
    warmup_scheduler = WarmUpLR(optimizer, iter_per_epoch * 1)
    checkpoint_path = 'checkpoint/'

    #input_tensor = torch.Tensor(12, 3, 32, 32).cpu()
    input_tensor = torch.Tensor(10, 64, 64, 3).cpu()

    # create checkpoint folder to save model
    if not os.path.exists(checkpoint_path):
        os.makedirs(checkpoint_path)
    checkpoint_path = os.path.join(checkpoint_path, '{net}-{epoch}-{type}.pth')
    best_acc = 0.0

    point_data_set = PointDataSet(X_train, Y_train)
    data_loader = DataLoader(dataset=point_data_set, batch_size=10, shuffle=True)
    data_test = PointDataSet(X_test, Y_test)
    test = DataLoader(dataset=data_test, batch_size=10, shuffle=True)

    for epoch in range(1, 3):
        if epoch > 1:
            train_scheduler.step(epoch)
        for batch_index, (x_batch, y_batch) in enumerate(data_loader):
            if epoch <= 1:
                warmup_scheduler.step()

            x_batch = x_batch.cpu()
            y_batch = y_batch.cpu()
            x_batch = Variable(x_batch.float())
            x_batch = x_batch.permute(0, 3, 1, 2)  # 将原来第1维变为0维
            y_batch = Variable(y_batch.long())
            y_batch = y_batch.squeeze(1)
            net.train()
            optimizer.zero_grad()
            outputs = net(x_batch)
            loss = loss_function(outputs, y_batch)
            loss.backward()
            optimizer.step()
            last_layer = list(net.children())[-1]
            print('Training Epoch: {epoch} Loss: {:0.4f} LR: {:0.6f}'.format(
                    loss.item(),
                    optimizer.param_groups[0]['lr'],
                    epoch=epoch
                ))

        for batch_index, (x_test, y_test) in enumerate(data_loader):
            net.eval()
            test_loss = 0.0  # cost function error
            correct = 0.0
            x_test = Variable(x_test.float())
            y_batch = Variable(y_batch.long())
            outputs = net(x_test)
            loss = loss_function(outputs, y_test)
            test_loss += loss.item()
            _, preds = outputs.max(1)
            correct += preds.eq(y_batch).sum()
            acc = correct.float() / 10
            print('Test set: Average loss: {:.4f}, Accuracy: {:.4f}'.format(
                test_loss / 10,
                acc
            ))


        # start to save best performance model after learning rate decay to 0.01
        if epoch > 120 and best_acc < acc:
            torch.save(net.state_dict(), checkpoint_path.format(net=net, epoch=epoch, type='best'))
            best_acc = acc
            continue

        if not epoch % 1:
            torch.save(net.state_dict(), checkpoint_path.format(net=net, epoch=epoch, type='regular'))