目录结构
- data 存储训练/测试数据
- test 测试集文件夹
- train 训练集文件夹
- alexnet-owt-4df8aa71.pth alexnet 预训练数据(网上可以下载)
- imagenet1000.json 训练数据分类标签
- imagenet_classnames.txt 训练数据分类标签
- tiger cat.jpg 测试用的随便找的
- results 训练结果
- src 主程序目录
- alexnet_inference.py 用预训练数据进行对猫狗的分类测试
- alexnet_visualizaton.py 用tensorboard包做的数据可视化,若您是pytorch1.7版本暂时有BUG无法运行
- train_alexnet.py 训练
alexnet_inference.py
该py文件为展示模型结构、测试分类结果,下面对代码的简单介绍:
def load_class_names(p_clsnames, p_clsnames_cn):
"""
加载标签名
:param p_clsnames:
:param p_clsnames_cn:
:return:
"""
with open(p_clsnames, "r") as f:
class_names = json.load(f)
with open(p_clsnames_cn, encoding='UTF-8') as f: # 设置文件对象
class_names_cn = f.readlines()
return class_names, class_names_cn
load_class_names 读取分类标签的,其中标签和训练/测试图片有一一对应的关系。
def get_model(path_state_dict, vis_model=False):
"""
创建模型,加载参数
:param path_state_dict:
:return:
"""
model = models.alexnet()
pretrained_state_dict = torch.load(path_state_dict)
model.load_state_dict(pretrained_state_dict)
model.eval()
if vis_model:
from torchsummary import summary
summary(model, input_size=(3, 224, 224), device="cpu")
model.to(device)
return model
获取模型,先定义alexnet模型。导入模型预训练数据,model.eval()停止对模型数据的更新操作
def process_img(path_img):
# hard code
norm_mean = [0.485, 0.456, 0.406]
norm_std = [0.229, 0.224, 0.225]
inference_transform = transforms.Compose([
transforms.Resize(256),
transforms.CenterCrop((224, 224)),
transforms.ToTensor(),
transforms.Normalize(norm_mean, norm_std),
])
# path --> img
img_rgb = Image.open(path_img).convert('RGB')
# img --> tensor
img_tensor = img_transform(img_rgb, inference_transform)
img_tensor.unsqueeze_(0) # chw --> bchw
img_tensor = img_tensor.to(device)
return img_tensor, img_rgb
对读入的图片进行预处理,对图片短边裁剪,中心化(裁剪图片亦 resize(224,224))、正则化否则会效果非常差,论文中有提到。
if __name__ == "__main__":
# config
path_state_dict = os.path.join(BASE_DIR, "..", "data", "alexnet-owt-4df8aa71.pth")
# path_img = os.path.join(BASE_DIR, "..", "data", "Golden Retriever from baidu.jpg")
path_img = os.path.join(BASE_DIR, "..", "data", "tiger cat.jpg")
path_classnames = os.path.join(BASE_DIR, "..", "data", "imagenet1000.json")
path_classnames_cn = os.path.join(BASE_DIR, "..", "data", "imagenet_classnames.txt")
# load class names
cls_n, cls_n_cn = load_class_names(path_classnames, path_classnames_cn)
# 1/5 load img
img_tensor, img_rgb = process_img(path_img)
# 2/5 load model
alexnet_model = get_model(path_state_dict, True)
# 3/5 inference tensor --> vector
with torch.no_grad():
time_tic = time.time()
outputs = alexnet_model(img_tensor)
time_toc = time.time()
# 4/5 index to class names
_, pred_int = torch.max(outputs.data, 1)
_, top5_idx = torch.topk(outputs.data, 5, dim=1)
# 如果想把CUDA tensor格式的数据改成numpy时,需要先将其转换成cpu float-tensor随后再转到numpy格式。 numpy不能读取CUDA tensor 需要将它转化为 CPU tensor
pred_idx = int(pred_int.cpu().numpy())
pred_str, pred_cn = cls_n[pred_idx], cls_n_cn[pred_idx]
print("img: {} is: {}\n{}".format(os.path.basename(path_img), pred_str, pred_cn))
print("time consuming:{:.2f}s".format(time_toc - time_tic))
# 5/5 visualization
plt.imshow(img_rgb)
plt.title("predict:{}".format(pred_str))
top5_num = top5_idx.cpu().numpy().squeeze()
text_str = [cls_n[t] for t in top5_num]
for idx in range(len(top5_num)):
plt.text(5, 15+idx*30, "top {}:{}".format(idx+1, text_str[idx]), bbox=dict(fc='yellow'))
plt.show()
分为五个部分:
1.加载图片
2.加载模型
3.用alexnet预测
4.预测结果分别为:pred_int top1 的结果,top5_idx top5的结果
5.可视化
效果展示:
展示网络层的具体细节用到了下面的代码:
from torchsummary import summary
summary(model, input_size=(3, 224, 224), device="cpu")
train_alexnet.py
不使用预训练参数,调大了训练模型的LR参数,直到30Epoch后模型才开始有一定的准确性,且收敛很慢。
数据库
kaggle:dogs vs cats
https://www.kaggle.com/c/dogs-vs-cats
预训练模型:
https://download.pytorch.org/models/alexnet-owt-4df8aa71.pth
源码(免费):
https://download.csdn.net/download/qq_33000453/13210953