Implementation:Open compass VLMEvalKit OlmOCRBench Evaluator
| Field | Value |
|---|---|
| source | VLMEvalKit |
| domain | Vision, Benchmarking, OCR, Document Parsing, Evaluation |
Overview
Evaluator module for the olmOCRBench document parsing and OCR benchmark in VLMEvalKit.
Description
This utility file does not define a dataset class. It provides evaluation logic for the olmOCRBench benchmark, implementing scoring and comparison functions for document parsing output. It handles evaluation of text extraction, mathematical formula recognition, and table structure conversion quality against ground-truth annotations.
Usage
Imported by vlmeval/dataset/olmOCRBench/olmocrbench.py to provide evaluation capabilities for the olmOCRBench dataset.
Code Reference
- Source:
vlmeval/dataset/olmOCRBench/evaluator.py, Lines: L1-467 - Import:
from vlmeval.dataset.olmOCRBench.evaluator import *
I/O Contract
| Direction | Description |
|---|---|
| Inputs | Predicted document parsing outputs and ground-truth annotations |
| Outputs | Evaluation scores for text, formula, and table parsing quality |
Related Pages
Page Connections
Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment