Implementation:Open compass VLMEvalKit MMReason Utils
| Field | Value |
|---|---|
| source | VLMEvalKit |
| domain | Vision, Benchmarking, Mathematical Reasoning, Utilities |
Overview
Utility module for MMReason evaluation in VLMEvalKit.
Description
This utility file does not define a dataset class. It provides evaluation helper functions for the MMReason mathematical reasoning benchmark including GPT-4 based answer extraction, score comparison, and per-discipline accuracy computation. Key functions include build_MMReason_gpt4_extract_prompt, build_MMReason_gpt4_score_prompt, and MMReason_acc.
Usage
Imported by related dataset modules in the VLMEvalKit evaluation pipeline.
Code Reference
- Source:
vlmeval/dataset/mmreason.py, Lines: L1-249 - Import:
from vlmeval.dataset.mmreason import *
I/O Contract
| Direction | Description |
|---|---|
| Inputs | Evaluation data, model predictions, ground-truth annotations |
| Outputs | Processed evaluation results and metric scores |
Related Pages
Page Connections
Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment