Fredham

为什么我使用readme.md里作者提供的Kinetics-400 pretrained model 执行‘./scripts/run_test.sh ./configs/k400/k400_test.yaml’命令就报这个错误： model = build_model(state_dict or model.state_dict(), joint=joint,tsm=tsm,T=T,dropout=dropout, emb_dropout=emb_dropout,pretrain=pretrain).to(device) File "/home/houyf22/lzu/ActionCLIP-master/clip/model.py", line 314, in build_model vision_width = state_dict["visual.layer1.0.conv1.weight"].shape[0] KeyError: 'visual.layer1.0.conv1.weight' 但是使用clip的预训练模型就不报错，是不是作者没有根据自己的预训练模型调整代码还是我漏了什么步骤？

How to build data?

2

The paper provides datasets to readers,while it doesn't tell us how to build data.For example,I want to imitate the format of MSVD to build my own data. I have raw...

使用自己的数据集

1

请问如何使用自己的数据集制作hdf5文件？