ValueError: loaded state dict contains a parameter group that doesn‘t match the size of optimizer‘s

state dict contains a parameter group that doesn‘t match the size of optimizer‘s"/>

ValueError: loaded state dict contains a parameter group that doesn‘t match the size of optimizer‘s

解决训练时报错：ValueError: loaded state dict contains a parameter group that doesn’t match the size of optimizer’s group
大致意思：加载状态 dict 包含一个与优化器组的大小不匹配的参数组
暂时解决：直接注释掉 optimizer.load_state_dict(ckpt[‘optimizer’])
因为
if cfg.BACKBONE.TRAIN_EPOCH == epoch:
logger.info(‘start training backbone.’)
optimizer, lr_scheduler = build_opt_lr(model.module, epoch)
logger.info(“model\n{}”.format(describe(model.module)))
当epoch=10 的时候，optimizer会重新生成。