Is there anything wrong with cache_llm.pkl?

[Weili-0234]

Hi, amazing work!
I just want to know if there's anything wrong with the file, "cache_llm.pkl"?
I was trying to run your code but I got some errors.
After checking line 91 of main.py and line 5 of utils_general.py, I think there may be something wrong with the file "cache_llm.pkl" you guys provided.
Could you provide some solutions to this? Thanks so much!

Below is the error message I got:

WARNING:root:Error getting from cache: b'["gpt-4-1106-preview", [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "\n Given a video that has 180 frames, the frames are decoded at 1 fps. Given the following descriptions of five uniformly sampled frames in the video:\n {'frame 1': '#C C pours the water from the bowl', 'frame 45': '#C C puts the sponge in the sink', 'frame 90': '#C C scrubs the plate with the sponge', 'frame 135': '#C C puts the soap bottle on the sink', 'frame 180': '#C C opens the soap bottle'}\n #C to denote the sentence is an action done by the camera wearer (the person who recorded the video while wearing a camera on their head).\n #O to denote that the sentence is an action done by someone other than the camera wearer.\n Please answer the following question: \n \\n Here is the question: Taking into account all the actions performed by c, what can you deduce about the primary objective and focus within the video content?\\nHere are the choices: 0. C is cooking. 1. C is doing laundry. 2. C is cleaning the kitchen. 3. C is cleaning dishes. 4. C is cleaning the bathroom.\\n \n Please think step-by-step and write the best answer index in Json format {'final_answer': 'xxx'}. Note that only one answer is returned for the question.\n "}]]'
Not hit cache ["gpt-4-1106-preview", [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "\n Given a video that has 180 frames, the frames are decoded at 1 fps. Given the following descriptions of five uniformly sampled frames in the video:\n {'frame 1': '#C C pours the water from the bowl', 'frame 45': '#C C puts the sponge in the sink', 'frame 90': '#C C scrubs the plate with the sponge', 'frame 135': '#C C puts the soap bottle on the sink', 'frame 180': '#C C opens the soap bottle'}\n #C to denote the sentence is an action done by the camera wearer (the person who recorded the video while wearing a camera on their head).\n #O to denote that the sentence is an action done by someone other than the camera wearer.\n Please answer the following question: \n \n Here is the question: Taking into account all the actions performed by c, what can you deduce about the primary objective and focus within the video content?\nHere are the choices: 0. C is cooking. 1. C is doing laundry. 2. C is cleaning the kitchen. 3. C is cleaning dishes. 4. C is cleaning the bathroom.\n \n Please think step-by-step and write the best answer index in Json format {'final_answer': 'xxx'}. Note that only one answer is returned for the question.\n "}]]
^CTraceback (most recent call last):