sunwhw

如果是参照官方的https://github.com/netease-youdao/QAnything/blob/qanything-python/docs/API.md#-urlhttpyour_host8777apilocal_doc_qalocal_doc_chat通过调用api的形式进行问答， 报这个错误的话，在取response那里改为 ```python res = response.json() # print(res) print(res['history'][-1][-1]) # 而不是 print(res['response']) 我看到源码的逻辑是会把这个真正的response存储到history里面，而“data: Done”应该是此次请求结束的标志，但可能这里没处理好，所以传错了，所以只需要从最新的history取答案即可

> which document should I download to get the narrations of the dataset? https://ego4d-data.org/docs/start-here/

So how to sovle it? I also get stuck with the problem.

I sovled! I changed the code just like ``` python def build_image_tower(image_tower_cfg, **kwargs): image_tower = getattr(image_tower_cfg, 'mm_image_tower', getattr(image_tower_cfg, 'image_tower', None)) is_absolute_path_exists = os.path.exists(image_tower) # if is_absolute_path_exists or image_tower.startswith("openai") or image_tower.startswith("laion"):...

Data: yes, and these data can be trained normally before modifying the model to support different frame. model: I also cut the len of data to 8, and it run...

Me too!

I've also encountered the same issue!

Is there anyone who still maintains this community? I try to input the whole true address, but it always keep "Approximate max length", so I only choose to esify the...