pytorch 保存和加载 Checkpoint 用于推理或继续训练

来源：bing 搜索结果 https://blog.csdn.net/Turbo_Come/article/details/105733552

保存和加载 Checkpoint 用于推理/继续训练

model = TheModelClass ( * args, ** kwargs)
optimizer = TheOptimizerClass ( * args, ** kwargs)
checkpoint = torch.load (PATH)
model.load_state_dict (checkpoint [ 'model_state_dict' ])
optimizer.load_state_dict (checkpoint [ 'optimizer_state_dict' ])
epoch = checkpoint [ 'epoch']
loss = checkpoint [ 'loss']
model.eval ()