地址链接:https://www.flyai.com/m/resnest200_d2-ca88e41f.pth
查看来源:https://download.openmmlab.com/pretrain/third_party/resnest200_d2-ca88e41f.pth
确定自己使用的框架并导入对应的库。导入库实现样例代码可参考 文档中心-预训练模型使用教程
在代码中实现加载预训练模型地址
import torchvisionfrom flyai.utils import remote_helper
path=remote_helper.get_remote_data("https://www.flyai.com/m/resnest200_d2-ca88e41f.pth")
model = torchvision.models.res2net101(pretrained = False) # model = torchvision.models.res2net101(pretrained = True) # 这行代码与上面等同,只不过一个是调用FlyAI提供的预训练模型地址,一个是外网的地址
model.load_state_dict(torch.load(path)# 将其中的层直接替换为我们需要的层即可
model.fc = nn.Linear(2048,200)
resnest200_d2-ca88e41f.pth
查看来源:https://download.openmmlab.com/pretrain/third_party/resnest200_d2-ca88e41f.pthresnest269-51ae5f19.pth
查看来源:https://s3.us-west-1.wasabisys.com/encoding/models/resnest269-51ae5f19.zipresnest200-d7fd712f.pth
查看来源:https://s3.us-west-1.wasabisys.com/encoding/models/resnest200-d7fd712f.zipresnest101-966fb78c.pth
查看来源:https://s3.us-west-1.wasabisys.com/encoding/models/resnest101-966fb78c.zip