這篇文章主要介紹了關於pytorch + visdom CNN處理自建圖片資料集的方法,有著一定的參考價值,現在分享給大家,有需要的朋友可以參考一下
環境
系統:win10
cpu:i7-6700HQ
gpu:gtx965m
python : 3.6
pytorch :0.3
資料下載
來源自Sasank Chilamkurthy 的教程; 資料:下載連結。
下載後解壓放到項目根目錄:
資料集為用來分類 螞蟻和蜜蜂。有大約120個訓練映像,每個類有75個驗證映像。
資料匯入
可以使用 torchvision.datasets.ImageFolder(root,transforms) 模組 可以將 圖片轉換為 tensor。
先定義transform:
ata_transforms = { 'train': transforms.Compose([ # 隨機切成224x224 大小圖片 統一圖片格式 transforms.RandomResizedCrop(224), # 映像翻轉 transforms.RandomHorizontalFlip(), # totensor 歸一化(0,255) >> (0,1) normalize channel=(channel-mean)/std transforms.ToTensor(), transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) ]), "val" : transforms.Compose([ # 圖片大小縮放 統一圖片格式 transforms.Resize(256), # 以中心裁剪 transforms.CenterCrop(224), transforms.ToTensor(), transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) ])}
匯入,載入資料:
data_dir = './hymenoptera_data'# trans dataimage_datasets = {x: datasets.ImageFolder(os.path.join(data_dir, x), data_transforms[x]) for x in ['train', 'val']}# load datadata_loaders = {x: DataLoader(image_datasets[x], batch_size=BATCH_SIZE, shuffle=True) for x in ['train', 'val']}data_sizes = {x: len(image_datasets[x]) for x in ['train', 'val']}class_names = image_datasets['train'].classesprint(data_sizes, class_names)
{'train': 244, 'val': 153} ['ants', 'bees']
訓練集 244圖片 , 測試集153圖片 。
可視化部分圖片看看,由於visdom支援tensor輸入 ,不用換成numpy,直接用tensor計算即可 :
inputs, classes = next(iter(data_loaders['val']))out = torchvision.utils.make_grid(inputs)inp = torch.transpose(out, 0, 2)mean = torch.FloatTensor([0.485, 0.456, 0.406])std = torch.FloatTensor([0.229, 0.224, 0.225])inp = std * inp + meaninp = torch.transpose(inp, 0, 2)viz.images(inp)
建立CNN
net 根據上一篇的處理cifar10的改了一下規格:
class CNN(nn.Module): def __init__(self, in_dim, n_class): super(CNN, self).__init__() self.cnn = nn.Sequential( nn.BatchNorm2d(in_dim), nn.ReLU(True), nn.Conv2d(in_dim, 16, 7), # 224 >> 218 nn.BatchNorm2d(16), nn.ReLU(inplace=True), nn.MaxPool2d(2, 2), # 218 >> 109 nn.ReLU(True), nn.Conv2d(16, 32, 5), # 105 nn.BatchNorm2d(32), nn.ReLU(True), nn.Conv2d(32, 64, 5), # 101 nn.BatchNorm2d(64), nn.ReLU(True), nn.Conv2d(64, 64, 3, 1, 1), nn.BatchNorm2d(64), nn.ReLU(True), nn.MaxPool2d(2, 2), # 101 >> 50 nn.Conv2d(64, 128, 3, 1, 1), # nn.BatchNorm2d(128), nn.ReLU(True), nn.MaxPool2d(3), # 50 >> 16 ) self.fc = nn.Sequential( nn.Linear(128*16*16, 120), nn.BatchNorm1d(120), nn.ReLU(True), nn.Linear(120, n_class)) def forward(self, x): out = self.cnn(x) out = self.fc(out.view(-1, 128*16*16)) return out# 輸入3層rgb ,輸出 分類 2 model = CNN(3, 2)
loss,最佳化函數:
line = viz.line(Y=np.arange(10))loss_f = nn.CrossEntropyLoss()optimizer = optim.SGD(model.parameters(), lr=LR, momentum=0.9)scheduler = optim.lr_scheduler.StepLR(optimizer, step_size=7, gamma=0.1)
參數:
BATCH_SIZE = 4LR = 0.001EPOCHS = 10
運行 10個 epoch 看看:
[9/10] train_loss:0.650|train_acc:0.639|test_loss:0.621|test_acc0.706[10/10] train_loss:0.645|train_acc:0.627|test_loss:0.654|test_acc0.686Training complete in 1m 16sBest val Acc: 0.712418
運行 20個看看:
[19/20] train_loss:0.592|train_acc:0.701|test_loss:0.563|test_acc0.712[20/20] train_loss:0.564|train_acc:0.721|test_loss:0.571|test_acc0.706Training complete in 2m 30sBest val Acc: 0.745098
準確率比較低:只有74.5%
我們使用models 裡的 resnet18 運行 10個epoch:
model = torchvision.models.resnet18(True)num_ftrs = model.fc.in_featuresmodel.fc = nn.Linear(num_ftrs, 2)
[9/10] train_loss:0.621|train_acc:0.652|test_loss:0.588|test_acc0.667[10/10] train_loss:0.610|train_acc:0.680|test_loss:0.561|test_acc0.667Training complete in 1m 24sBest val Acc: 0.686275
效果也很一般,想要短時間內就訓練出效果很好的models,我們可以下載訓練好的state,在此基礎上訓練:
model = torchvision.models.resnet18(pretrained=True)num_ftrs = model.fc.in_featuresmodel.fc = nn.Linear(num_ftrs, 2)
[9/10] train_loss:0.308|train_acc:0.877|test_loss:0.160|test_acc0.941[10/10] train_loss:0.267|train_acc:0.885|test_loss:0.148|test_acc0.954Training complete in 1m 25sBest val Acc: 0.954248
10個epoch直接的到95%的準確率。