Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/NVlabs/OmniVinci
/ functions
Functions
10 in github.com/NVlabs/OmniVinci
⨍
Functions
10
◇
Types & classes
1
↓ 2 callers
Method
generate_response
Generate a response from the model given a video and text prompt. Args: video_path (str): Path to the video file
example_infer.py:141
↓ 2 callers
Method
validate_paths
Validate that required paths exist.
example_infer.py:61
↓ 1 callers
Method
_print_model_info
Print useful information about the loaded model.
example_infer.py:107
↓ 1 callers
Function
add_to_sys_path_direct
Add model path directly to sys.path
example_infer.py:31
↓ 1 callers
Method
batch_generate
Generate responses for multiple video-text pairs. Args: video_text_pairs (List[tuple]): List of (video_path, tex
example_infer.py:250
↓ 1 callers
Method
create_conversation
Create a conversation format for the model. Args: video_path (str): Path to the video file text_prom
example_infer.py:121
↓ 1 callers
Function
forward_inference
Run inference on video with text prompt.
example_mini_video.py:78
↓ 1 callers
Method
load_model
Load the model, processor, and config with error handling.
example_infer.py:73
↓ 1 callers
Function
main
Main function demonstrating usage of the NVOmni model.
example_infer.py:277
Method
__init__
Initialize the NVOmni model for video inference. Args: model_path (str): Path to the model directory
example_infer.py:42