地址链接:https://www.flyai.com/m/pre_efficientnetv2-m.pth
查看来源:https://storage.googleapis.com/cloud-tpu-checkpoints/efficientnet/v2/efficientnetv2-m.tgz
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 efficientnetv2_m
import torch
from flyai.utilsimport remote_helper
path = remote_helper.get_remote_data('https://www.flyai.com/m/pre_efficientnetv2-m.pth')
self.net = efficientnetv2_m(num_classes)
self.net.to(device)
weights_dict = torch.load(path, map_location=device)
load_weights_dict = {k: v for k, v in weights_dict.items() if self.net.state_dict()[k].numel() == v.numel()}
self.net.load_state_dict(load_weights_dict, strict=False)
tf_efficientnetv2_s_21k-6337ad01.pth
查看来源:https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_s_21k-6337ad01.pthtf_efficientnetv2_l_21k-91a19ec9.pth
查看来源:https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-effv2-weights/tf_efficientnetv2_l_21k-91a19ec9.pthmobilenetv3-large-1cd25616.pth
查看来源:https://github.com/bubbliiiing/mobilenet-yolov4-lite-pytorch#%E6%96%87%E4%BB%B6%E4%B8%8B%E8%BD%BDinception_v3_google-0cc3c7bd.pth
查看来源:https://download.pytorch.org/models/inception_v3_google-0cc3c7bd.pthdeit_base_patch16_224-b5f2ef4d.pth
查看来源:https://dl.fbaipublicfiles.com/deit/deit_base_patch16_224-b5f2ef4d.pthswin_large_patch4_window12_384_22k.pth
查看来源:https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_large_patch4_window12_384_22k.pth