tianguang

有什么模型可以只输入视频和音频，不需要输入caption的吗

4

如题

about the parameters of train.py

1

Hello, thanks for sharing the wonderful code! Could you please tell me how to configure the parameters of the train.py file to obtain the effect of the published model, or...

1、Why in 224_v2_llama2_video_stage_3.yaml the config is llama_model: "meta-llama/Meta-Llama-3-8B-Instruct".While in stage_2, it is "meta-llama/Llama-2-7b-chat-hf". 2、Why the task is "image_text_pretrain" in stead of "fituning " in stage 3. Thanks.

RuntimeError: Parent directory ./Video-ChatGPT_7B-1.1_Checkpoints/checkpoint-3000 does not exist.

5

When the steps run tosave_steps, the above error will be reported

Conductor选择出的专家总是错误

1

如图是conductor.py的调试结果，预测的专家总是存在***，导致每次只能随机选择。