Skip to content

GoldFish LLaMa 2 Inference #35

@insafim

Description

@insafim

{'default': 'configs/datasets/video_chatgpt/default.yaml'}
{'default': 'path to the config file'}
using openai: True
Initialization Model

model arch mini_gpt4_llama_v2
model cls <class 'minigpt4.models.mini_gpt4_llama_v2.MiniGPT4_Video'>
dataset name video_chatgpt
Error setting attribute device with value cuda
Llama model
token pooling True
vit precision fp16
freeze the vision encoder
Loading VIT Done
Loading LLAMA
self.low_resource True
Loading checkpoint shards: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:09<00:00, 4.85s/it]
trainable params: 33,554,432 || all params: 6,771,970,048 || trainable%: 0.4955
Loading LLAMA Done
Load Minigpt-4-LLM Checkpoint: /share/data/drive_4/insaf/checkpoints/MiniGPT4-Video/checkpoints/video_captioning_llama_checkpoint_last.pth
{'name': 'blip2_image_train', 'image_size': 224}
Initialization Finished
Loading checkpoint shards: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:09<00:00, 2.42s/it]
Time taken to load model: 73.33053708076477
Video name: 203
Video duration: 480.00 seconds
Long video
External memory is ready
Loading embeddings from pkl file
Traceback (most recent call last):
File "/share/data/drive_4/insaf/MiniGPT4-video/goldfish_inference.py", line 59, in
pred=process_video(processed_video_path, args.add_subtitles, args.question)
File "/share/data/drive_4/insaf/MiniGPT4-video/goldfish_inference.py", line 40, in process_video
result = goldfish_lv.inference(video_path, has_subtitles, instruction,number_of_neighbours)
File "/share/data/drive_4/insaf/MiniGPT4-video/goldfish_lv.py", line 560, in inference
related_information=self.get_related_context(external_memory,related_context_keys)
File "/share/data/drive_4/insaf/MiniGPT4-video/goldfish_lv.py", line 510, in get_related_context
most_related_clips=self.get_most_related_clips(related_context_keys)
File "/share/data/drive_4/insaf/MiniGPT4-video/goldfish_lv.py", line 506, in get_most_related_clips
assert len(most_related_clips)!=0, f"No related clips found {related_context_keys}"
AssertionError: No related clips found []2

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions