天池CV入门赛SVHN代码分析

1. Package

# -*- coding: utf-8 -*-                                                                          
import os, sys, glob, shutil, json
os.environ["CUDA_VISIBLE_DEVICES"] = '0'

import cv2
import numpy as np
from PIL import Image

from tqdm import tqdm, tqdm_notebook

%pylab inline

import torch
torch.manual_seed(0)
torch.backends.cudnn.deterministic = False
torch.backends.cudnn.benchmark = True

import torchvision.models as models
import torchvision.transforms as transforms
import torchvision.datasets as datasets
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torch.autograd import Variable
from torch.utils.data.dataset import Dataset

2. Data

定义数据集

def __init__(self, loader=default_loader):

这个里面一般要初始化一个loader(代码见上面),一个images_path的列表，一个target的列表

def __getitem__(self, index):

这里就是在给你一个index的时候，你返回一个图片的tensor和target的tensor,使用了loader方法，经过 归一化，剪裁，类型转化，从图像变成tensor

def __len__(self):

return所有数据的个数

这三个综合起来看呢，其实就是输入所有数据的长度，它每次给你返回一个shuffle过的index,以这个方式遍历数据集，通过 getitem(self, index)返回一组你要的（input,target）

class SVHNDataset(Dataset):
    def __init__(self, img_path, img_label, transform=None):
        self.img_path = img_path
        self.img_label = img_label
        if transform is not None:
            self.transform = transform
        else:
            self.transform = None

    def __getitem__(self, index):
        img = Image.open(self.img_path[index]).convert('RGB')

        if self.transform is not None:
            img = self.transform(img)

        lbl = np.array(self.img_label[index], dtype=np.int)
        lbl = list(lbl)  + (5 - len(lbl)) * [10]
        return img, torch.from_numpy(np.array(lbl[:5]))

    def __len__(self):
        return len(self.img_path)

# 训练集
# glob.glob(): 匹配符合条件的所有文件，并将其以list的形式返回
# train_path: ['./data/mchar_train/000000.png',...]
train_path = glob.glob(`'./data/mchar_train/*.png'`)
train_path.sort()
train_json = json.load(open(`'./data/mchar_train.json'`))
train_label = [train_json[x]['label'] for x in train_json]
print(len(train_path), len(train_label))
# train data:
# 30000 30000

# 验证集
val_path = glob.glob('../../../dataset/tianchi_SVHN/val/*.png')
val_path.sort()
val_json = json.load(open('../../../dataset/tianchi_SVHN/val.json'))
val_label = [val_json[x]['label'] for x in val_json]
print(len(val_path), len(val_label))
# val data:
# 10000 10000

# 测试集
test_path = glob.glob('./data/mchar_test_a/*.png')
test_path.sort()
test_label = [[1]] * len(test_path)
print(len(test_path), len(test_label))

定义读取数据dataloader

# Dataset：对数据集的封装，提供索引方式的对数据样本进行读取
# DataLoder：对Dataset进行封装，提供批量读取的迭代读取
# 在加入DataLoder后，数据按照批次获取，每批次调用Dataset读取单个样本进行拼接。

# train
# SVHNDataset参数：img_path, img_label, transform
train_loader = torch.utils.data.DataLoader(
    SVHNDataset(train_path, train_label,
                transforms.Compose([
                    # transforms.Compose: 
                    # 传入参数为一个列表，列表中的元素是对数据进行变换的操作
                    transforms.Resize((64, 128)),
                    transforms.RandomCrop((60, 120)),
                    transforms.ToTensor(),
                    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ])),
    batch_size=40,
    shuffle=True,
    num_workers=10,
)
# val
val_loader = torch.utils.data.DataLoader(
    SVHNDataset(val_path, val_label,
                transforms.Compose([
                    transforms.Resize((60, 120)),
                    transforms.ToTensor(),
                    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ])),
    batch_size=40,
    shuffle=False,
    num_workers=10,
)
# test
test_loader = torch.utils.data.DataLoader(
    SVHNDataset(test_path, test_label,
                transforms.Compose([
                    transforms.Resize((70, 140)),
                    transforms.ToTensor(),
                    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ])),
    batch_size=40,
    shuffle=False,
    num_workers=10,
)

DataLoader

Model

class SVHN_Model1(nn.Module):
    def __init__(self):
        super(SVHN_Model1, self).__init__()
        # 使用resnet18网络
        model_conv = models.resnet18(pretrained=True)
        
        # 将resnet的平均池化设置为自适应平均池化*
        # nn.AdaptiveAvgPool2d(output_size):
        # 将输出尺寸指定为output_size，通道数前后不发生变化。
        model_conv.avgpool = nn.AdaptiveAvgPool2d(1)
        model_conv = nn.Sequential(*list(model_conv.children())[:-1])
        self.cnn = model_conv
        
        self.fc1 = nn.Linear(512, 11)
        self.fc2 = nn.Linear(512, 11)
        self.fc3 = nn.Linear(512, 11)
        self.fc4 = nn.Linear(512, 11)
        self.fc5 = nn.Linear(512, 11)
    
    def forward(self, img):        
        feat = self.cnn(img)
        # print(feat.shape)
        feat = feat.view(feat.shape[0], -1)
        c1 = self.fc1(feat)
        c2 = self.fc2(feat)
        c3 = self.fc3(feat)
        c4 = self.fc4(feat)
        c5 = self.fc5(feat)
        return c1, c2, c3, c4, c5

*nn.AdaptiveAvgPool2d()

• 参数形式为 output_size或H*W

m = nn.AdaptiveAvgPool2d((3,7))
input = torch.randn(1, 64, 8, 9)
print（m（input））
#torch.Size([1, 64, 3, 7])

m = nn.AdaptiveAvgPool2d(7)
input = torch.randn(1, 64, 10, 9)
print（m（input））
#torch.Size([1, 64, 7, 7])

m = nn.AdaptiveAvgPool2d((None, 3))
input = torch.randn(1, 64, 10, 9)
print（m（input））
# torch.Size([1, 64, 10, 3])

Train()

def train(train_loader, model, criterion, optimizer):
    # 切换模型为训练模式
    model.train()
    train_loss = []

    for i, (input, target) in enumerate(train_loader):
        if use_cuda:
            input = input.cuda()
            target = target.cuda()

        c0, c1, c2, c3, c4 = model(input)
        loss = criterion(c0, target[:, 0]) + \
                criterion(c1, target[:, 1]) + \
                criterion(c2, target[:, 2]) + \
                criterion(c3, target[:, 3]) + \
                criterion(c4, target[:, 4])

        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        train_loss.append(loss.item())
    return np.mean(train_loss)

……

Predict()

def predict(test_loader, model, tta=10):
    model.eval()
    test_pred_tta = None
    
    # TTA 次数
    for _ in range(tta):
        test_pred = []
    
        with torch.no_grad():
            for i, (input, target) in enumerate(test_loader):
                if use_cuda:
                    input = input.cuda()
                
                c0, c1, c2, c3, c4 = model(input)
                if use_cuda:
                    output = np.concatenate([
                        c0.data.cpu().numpy(), 
                        c1.data.cpu().numpy(),
                        c2.data.cpu().numpy(), 
                        c3.data.cpu().numpy(),
                        c4.data.cpu().numpy()], axis=1)
                else:
                    output = np.concatenate([
                        c0.data.numpy(), 
                        c1.data.numpy(),
                        c2.data.numpy(), 
                        c3.data.numpy(),
                        c4.data.numpy()], axis=1)
                
                test_pred.append(output)
        
        test_pred = np.vstack(test_pred)
        if test_pred_tta is None:
            test_pred_tta = test_pred
        else:
            test_pred_tta += test_pred
    
    return test_pred_tta

Run

model = SVHN_Model1()
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(model.parameters(), 0.001)

use_cuda = True
if use_cuda:
    model = model.cuda()
best_loss = 1000.0

for epoch in range(10):
    train_loss = train(train_loader, model, criterion, optimizer)
    val_loss = validate(val_loader, model, criterion)
    
    val_label = [''.join(map(str, x)) for x in val_loader.dataset.img_label]
    val_predict_label = predict(val_loader, model, 1)
    val_predict_label = np.vstack([
        val_predict_label[:, :11].argmax(1),
        val_predict_label[:, 11:22].argmax(1),
        val_predict_label[:, 22:33].argmax(1),
        val_predict_label[:, 33:44].argmax(1),
        val_predict_label[:, 44:55].argmax(1),
    ]).T
    val_label_pred = []
    for x in val_predict_label:
        val_label_pred.append(''.join(map(str, x[x!=10])))
    
    val_char_acc = np.mean(np.array(val_label_pred) == np.array(val_label))
    
    print('Epoch: {0}, Train loss: {1} \t Val loss: {2} \t Val Acc: {3}'.format(epoch, train_loss, val_loss, val_char_acc))

    # 记录下验证集精度
    if val_loss < best_loss:
        best_loss = val_loss
        # print('Find better model in Epoch {0}, saving model.'.format(epoch))
        torch.save(model.state_dict(), './model.pt')

# 加载保存的最优模型
model.load_state_dict(torch.load('./model.pt'))

test_predict_label = predict(test_loader, model, 1)
print(test_predict_label.shape)

test_label = [''.join(map(str, x)) for x in test_loader.dataset.img_label]
test_predict_label = np.vstack([
    test_predict_label[:, :11].argmax(1),
    test_predict_label[:, 11:22].argmax(1),
    test_predict_label[:, 22:33].argmax(1),
    test_predict_label[:, 33:44].argmax(1),
    test_predict_label[:, 44:55].argmax(1),
]).T

test_label_pred = []
for x in test_predict_label:
    test_label_pred.append(''.join(map(str, x[x!=10])))

• 按照题目要求格式存储预测结果

import pandas as pd
df_submit = pd.read_csv('./data/mchar_sample_submit_A.csv')
df_submit['file_code'] = test_label_pred
df_submit.to_csv('submit.csv', index=None)

---

代码来源：DataWhale