Implementation:Open compass VLMEvalKit MaCBench

Field	Value
source	VLMEvalKit
domain	Vision, Benchmarking, Multi-modal Comprehension

Overview

Benchmark dataset implementation for MaCBench evaluation in VLMEvalKit.

Description

MaCBench inherits from ImageBaseDataset and implements the MaCBench multi-modal comprehension benchmark. The TYPE field is set to 'MaCBench'. It supports interleaved image-text prompts with image placeholders embedded in question text.

Usage

Registered in vlmeval/dataset/__init__.py and invoked through build_dataset() by benchmark name.

Code Reference

Source: vlmeval/dataset/macbench.py, Lines: L1-292
Import: from vlmeval.dataset.macbench import MaCBench

Signature:

class MaCBench(ImageBaseDataset):
    TYPE = 'MaCBench'
    DATASET_URL = {...}
    DATASET_MD5 = {...}
    ...

I/O Contract

Direction	Description
Inputs	TSV dataset file with image/video paths and questions
Outputs	Evaluation results DataFrame with scores per category

Usage Examples

from vlmeval.dataset import build_dataset
dataset = build_dataset('MaCBench')

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment