# spraakbanken/multiged-2023 architecture

> Architecture diagram of the GitHub repository spraakbanken/multiged-2023, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-04). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/spraakbanken/multiged-2023
- Repository: https://github.com/spraakbanken/multiged-2023

## Overview

MultiGED-2023 is a shared-task data and evaluation repository for token-level grammatical error detection in Czech, English, German, Italian, and Swedish. Participants obtain annotated corpus splits, develop systems that label tokens correct or incorrect, and evaluate hypothesis TSVs against references with the included scorer; German M2 annotations can also be converted to a tabular format. The README documents the task, datasets, and official CodaLab evaluation. The sampled scripts show conversion and scoring behavior; dataset internals and participant systems are not part of this repository.

Main entry points: [`eval.py`](https://github.com/spraakbanken/multiged-2023/blob/main/eval.py) (Token scorer), [`german`](https://github.com/spraakbanken/multiged-2023/tree/main/german) (German Falko-MERLIN), [`czech`](https://github.com/spraakbanken/multiged-2023/tree/main/czech) (Czech GECCC), [`english`](https://github.com/spraakbanken/multiged-2023/tree/main/english) (English corpora)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_corpora["Language corpora"]
  node_czech[("Czech GECCC")]
  node_english[("English corpora")]
  node_german[("German Falko-MERLIN")]
  node_italian[("Italian MERLIN")]
  node_swedish[("Swedish SweLL")]
end

subgraph group_utilities["Task utilities"]
  node_converter["M2 converter"]
  node_scorer["Token scorer<br/>[eval.py]"]
end

node_participant(("Participant"))
node_hypothesis["Hypothesis TSV"]
node_codalab["CodaLab evaluation"]

node_participant -->|"gets data"| node_czech
node_participant -->|"gets data"| node_english
node_participant -->|"gets data"| node_german
node_participant -->|"gets data"| node_italian
node_participant -->|"gets data"| node_swedish
node_german -->|"M2 input"| node_converter
node_converter -->|"writes tabular output"| node_participant
node_participant -.->|"produces predictions"| node_hypothesis
node_hypothesis -->|"hypothesis input"| node_scorer
node_czech -.->|"reference input"| node_scorer
node_english -.->|"reference input"| node_scorer
node_german -.->|"reference input"| node_scorer
node_italian -.->|"reference input"| node_scorer
node_swedish -.->|"reference input"| node_scorer
node_participant -->|"submits system"| node_codalab

click node_czech "https://github.com/spraakbanken/multiged-2023/tree/main/czech"
click node_english "https://github.com/spraakbanken/multiged-2023/tree/main/english"
click node_german "https://github.com/spraakbanken/multiged-2023/tree/main/german"
click node_italian "https://github.com/spraakbanken/multiged-2023/tree/main/italian"
click node_swedish "https://github.com/spraakbanken/multiged-2023/tree/main/swedish"
click node_converter "https://github.com/spraakbanken/multiged-2023/blob/main/m2_to_conll_conversion-script.py"
click node_scorer "https://github.com/spraakbanken/multiged-2023/blob/main/eval.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_czech,node_english,node_german,node_italian,node_swedish toneBlue
class node_converter,node_scorer toneAmber
class node_participant,node_hypothesis,node_codalab toneIndigo
```

## Components

### Language corpora

- [Czech GECCC](https://github.com/spraakbanken/multiged-2023/tree/main/czech): `czech`
- [English corpora](https://github.com/spraakbanken/multiged-2023/tree/main/english): `english`
- [German Falko-MERLIN](https://github.com/spraakbanken/multiged-2023/tree/main/german): `german`
- [Italian MERLIN](https://github.com/spraakbanken/multiged-2023/tree/main/italian): `italian`
- [Swedish SweLL](https://github.com/spraakbanken/multiged-2023/tree/main/swedish): `swedish`

### Task utilities

- [M2 converter](https://github.com/spraakbanken/multiged-2023/blob/main/m2_to_conll_conversion-script.py): `m2_to_conll_conversion-script.py`
- [Token scorer](https://github.com/spraakbanken/multiged-2023/blob/main/eval.py): `eval.py`

### Other components

- **Participant**
- **Hypothesis TSV**
- **CodaLab evaluation**

## Connections

- Participant → Czech GECCC: gets data (evidence: `README.md`)
- Participant → English corpora: gets data (evidence: `README.md`)
- Participant → German Falko-MERLIN: gets data (evidence: `README.md`)
- Participant → Italian MERLIN: gets data (evidence: `README.md`)
- Participant → Swedish SweLL: gets data (evidence: `README.md`)
- German Falko-MERLIN → M2 converter: M2 input (evidence: `m2_to_conll_conversion-script.py`)
- M2 converter → Participant: writes tabular output (evidence: `m2_to_conll_conversion-script.py`)
- Participant → Hypothesis TSV: produces predictions
- Hypothesis TSV → Token scorer: hypothesis input (evidence: `eval.py`)
- Czech GECCC → Token scorer: reference input
- English corpora → Token scorer: reference input
- German Falko-MERLIN → Token scorer: reference input
- Italian MERLIN → Token scorer: reference input
- Swedish SweLL → Token scorer: reference input
- Participant → CodaLab evaluation: submits system (evidence: `README.md`)
