Jump to content

Connect SuperML | Leeroopedia MCP: Equip your AI agents with best practices, code verification, and debugging knowledge. Powered by Leeroo — building Organizational Superintelligence. Contact us at founders@leeroo.com.

Implementation:Open compass VLMEvalKit CGAVCounting Utils

From Leeroopedia
Revision as of 13:28, 16 February 2026 by Admin (talk | contribs) (Auto-imported from implementations/Open_compass_VLMEvalKit_CGAVCounting_Utils.md)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Field Value
source VLMEvalKit
domain Vision, Benchmarking, Video Counting Utilities

Overview

Utility module providing scoring, timestamp parsing, and evaluation helper functions for the CG-AV-Counting benchmark in VLMEvalKit.

Description

This utility file does not define a dataset class. It provides helper functions including rating_func for computing evaluation metrics (accuracy, OBOA, MAE, RMSE) across different task modes (long_acc, ref_acc, clue_acc), timestamp conversion utilities, and timeout handling for the CGAVCounting benchmark evaluation pipeline.

Usage

Imported by vlmeval/dataset/CGAVCounting/cg_av_counting.py via wildcard import to support the CGAVCounting dataset evaluation.

Code Reference

  • Source: vlmeval/dataset/CGAVCounting/utils.py, Lines: L1-422
  • Import: from vlmeval.dataset.CGAVCounting.utils import *

Key Functions:

def rating_func(data_path):
    ...

def get_timestampes(frame_indices, fps):
    ...

def time_str_to_seconds(time_str: str) -> float:
    ...

I/O Contract

Direction Description
Inputs Data path to evaluation results, frame indices and FPS values
Outputs Rating dictionaries with metric scores, formatted timestamp strings

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment