【猫狗数据集】对一张张图像进行预测（而不是测试集）

import cv2
import matplotlib.pyplot as plt
import glob

# 使用matplotlib展示多张图片
def matplotlib_multi_pic1():
  i=0
  for img in glob.glob('/content/drive/My Drive/colab notebooks/image/*.jpg'):
    img_name=img.split("/")[-1]
    img = cv2.imread(img)
    title=img_name
    #行，列，索引
    plt.subplot(3,3,i+1)
    plt.imshow(img)
    plt.title(title,fontsize=8)
    plt.xticks([])
    plt.yticks([])
    i+=1
  plt.show()
matplotlib_multi_pic1()

import torchvision
import sys
import torch
import torch.nn as nn
from PIL import Image
sys.path.append("/content/drive/My Drive/colab notebooks")
import glob
import numpy as np
import torchvision.transforms as transforms

device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
model=torchvision.models.resnet18(pretrained=False)
model.fc = nn.Linear(model.fc.in_features,2,bias=False)
model.to(device)
model.eval()
save_path="/content/drive/My Drive/colab notebooks/output/resnet18_best.t7" 
checkpoint = torch.load(save_path)
model.load_state_dict(checkpoint['model'])
print("当前模型准确率为：",checkpoint["epoch_acc"])
images_path="/content/drive/My Drive/colab notebooks/image/"
transform = transforms.Compose([transforms.Resize((224,224))])
def predict():
  true_labels=[]
  output_labels=[]
  for image in glob.glob(images_path+"/*.jpg"):
    print(image)
    if "cat" in image.split("/")[-1]:
      tmp=0
    else:
      tmp=1
    true_labels.append(tmp)
    image=Image.open(image)
    image=image.resize((224,224))
    tensor=torch.from_numpy(np.asarray(image)).permute(2,0,1).float()/255.0
    print(tensor.shape)
    tensor=tensor.reshape((1,3,224,224))
    tensor=tensor.to(device)
    #print(tensor.shape)
    output=model(tensor)
    print(output)
    _, pred = torch.max(output.data,1)
    output_labels.append(pred.item())
  return true_labels,output_labels

true_labels,output_labels=predict()
print("正确的标签是：")
print(true_labels)
print("预测的标签是：")
print(output_labels)