Implementation:Open compass VLMEvalKit GSM8KVDataset

Field	Value
source	VLMEvalKit
domain	Vision, Benchmarking, Mathematical Reasoning, Visual

Overview

Benchmark dataset implementation for GSM8K-V evaluation in VLMEvalKit.

Description

GSM8KVDataset inherits from ImageBaseDataset and implements the GSM8K-V benchmark for visual mathematical reasoning. The TYPE field is set to 'VQA'. It supports multiple evaluation modes (text_only, visual_explicit, visual_implicit, all) configured via environment variable.

Usage

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

Code Reference

Source: vlmeval/dataset/gsm8k_v.py, Lines: L1-177
Import: from vlmeval.dataset.gsm8k_v import GSM8KVDataset

Signature:

class GSM8KVDataset(ImageBaseDataset):
    TYPE = 'VQA'
    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('GSM8K-V')

Related Pages

Page Connections

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

Principle

Implementation

Heuristic

Environment