Implementation:Open compass VLMEvalKit Spatial457
Appearance
| Field | Value |
|---|---|
| source | VLMEvalKit |
| domain | Vision, Benchmarking, Spatial Reasoning |
Overview
Benchmark dataset implementation for Spatial457 evaluation in VLMEvalKit.
Description
Spatial457 inherits from ImageBaseDataset and implements the Spatial457 benchmark for spatial reasoning evaluation. The TYPE field is set to 'VQA'. It includes a ROBUST mode where non-conforming responses are treated as full answers.
Usage
Registered in vlmeval/dataset/__init__.py and invoked through build_dataset() by benchmark name.
Code Reference
- Source:
vlmeval/dataset/spatial457.py, Lines: L1-212 - Import:
from vlmeval.dataset.spatial457 import Spatial457
Signature:
class Spatial457(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('Spatial457')
Related Pages
Page Connections
Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment