Request for InternVideo2-stage2 6b model

[dipta007]

Hello, Is there any approximate time when the 6b model will be available that had been used to get the rank 1 result on the MSR-VTT leaderboard?

[leexinhao]

It's available at https://huggingface.co/OpenGVLab/InternVideo2-Stage2_6B-224p-f4

[dipta007]

Thanks @leexinhao , any idea when the script or config will be available for MSRVTT?

[dipta007]

Also, the weight gives error:

import torch
state_dict = torch.load("data/models/InternVid2/internvideo2-s2_6b-224p-f4.pt", map_location='cpu')

Error:

<stdin>:1: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/sroydip1/.conda/envs/multivent/lib/python3.10/site-packages/torch/serialization.py", line 1360, in load
    return _load(
  File "/home/sroydip1/.conda/envs/multivent/lib/python3.10/site-packages/torch/serialization.py", line 1848, in _load
    result = unpickler.load()
  File "/home/sroydip1/.conda/envs/multivent/lib/python3.10/site-packages/torch/serialization.py", line 1837, in find_class
    return super().find_class(mod_name, name)
ModuleNotFoundError: No module named 'utils'

[leexinhao]

You should ref to https://github.com/OpenGVLab/InternVideo/blob/main/InternVideo2/multi_modality/scripts/evaluation/stage2/zero_shot/1B/eval_msrvtt.sh to test it.

[leexinhao]

Also, the weight gives error:

import torch
state_dict = torch.load("data/models/InternVid2/internvideo2-s2_6b-224p-f4.pt", map_location='cpu')

Error:

<stdin>:1: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/sroydip1/.conda/envs/multivent/lib/python3.10/site-packages/torch/serialization.py", line 1360, in load
    return _load(
  File "/home/sroydip1/.conda/envs/multivent/lib/python3.10/site-packages/torch/serialization.py", line 1848, in _load
    result = unpickler.load()
  File "/home/sroydip1/.conda/envs/multivent/lib/python3.10/site-packages/torch/serialization.py", line 1837, in find_class
    return super().find_class(mod_name, name)
ModuleNotFoundError: No module named 'utils'

You should load it in our codebase dir.

[dipta007]

You should ref to https://github.com/OpenGVLab/InternVideo/blob/main/InternVideo2/multi_modality/scripts/evaluation/stage2/zero_shot/1B/eval_msrvtt.sh to test it.

I will try but should the config be totally same as 6b? 🤔🤔