# divyajyoti019/nlp-cross-lingual-project architecture

> Architecture diagram of the GitHub repository divyajyoti019/nlp-cross-lingual-project, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-19). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/divyajyoti019/nlp-cross-lingual-project
- Repository: https://github.com/divyajyoti019/nlp-cross-lingual-project

## Overview

IndicRAG-QA is a multilingual education-policy question-answering demo. Users submit questions through the browser app or CLI; the system retrieves and ranks policy passages, applies an answerability threshold, then returns an evidence-grounded answer or abstains, with citations. Document ingestion creates searchable passages, while evaluation measures retrieval, answer quality, and abstention. The graph also preserves documented browser capabilities and optional integrations. Some UI wiring and evaluation details are unsampled, so those relationships are shown only where the README or excerpts support them.

Main entry points: [`src/indicrag/pipeline.py`](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/pipeline.py) (QA orchestration), [`app.py`](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/app.py) (Browser app), [`src/indicrag/cli.py`](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/cli.py) (CLI commands), [`src/indicrag/retrieval.py`](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/retrieval.py) (Passage retrieval)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_interfaces["User interfaces"]
  node_browser["Browser app<br/>[app.py]"]
  node_cli["CLI commands<br/>[cli.py]"]
  node_compare["Method comparison<br/>[app.py]"]
  node_explorer["Corpus explorer<br/>[app.py]"]
  node_upload["Document upload<br/>[app.py]"]
end

subgraph group_qa["Question answering"]
  node_engine["QA orchestration<br/>[pipeline.py]"]
  node_retriever["Passage retrieval<br/>[retrieval.py]"]
  node_answering["Answer generation<br/>[generation.py]"]
  node_passage_schema["Passage schema<br/>[schema.py]"]
end

subgraph group_corpus["Corpus and ingestion"]
  node_ingest["Document ingestion<br/>[ingest.py]"]
  node_starter["Starter passages<br/>[passages.jsonl]"]
  node_extended["Expanded passages"]
end

subgraph group_assessment["Evaluation"]
  node_evaluation["Evaluation metrics<br/>[evaluation.py]"]
  node_qa_data["QA evaluation set<br/>[qa.jsonl]"]
end

node_user(("Questioner"))
node_ollama["Ollama service"]

node_user -->|"asks via"| node_browser
node_user -->|"asks via"| node_cli
node_browser -->|"submits questions"| node_engine
node_cli -->|"runs ask"| node_engine
node_engine -->|"searches passages"| node_retriever
node_engine -->|"generates answer"| node_answering
node_engine -->|"uses passage data"| node_passage_schema
node_retriever -->|"ranks passages"| node_passage_schema
node_answering -.->|"calls when enabled"| node_ollama
node_browser -->|"offers comparison"| node_compare
node_compare -.->|"compares methods"| node_retriever
node_browser -->|"offers search"| node_explorer
node_browser -->|"offers uploads"| node_upload
node_upload -.->|"ingests documents"| node_ingest
node_cli -->|"runs ingestion"| node_ingest
node_ingest -->|"creates passages"| node_passage_schema
node_starter -.->|"loads corpus"| node_engine
node_extended -.->|"loads optionally"| node_engine
node_cli -->|"runs evaluation"| node_evaluation
node_evaluation -->|"scores retrieval"| node_retriever
node_evaluation -.->|"reads questions"| node_qa_data

click node_browser "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/app.py"
click node_cli "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/cli.py"
click node_compare "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/app.py"
click node_explorer "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/app.py"
click node_upload "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/app.py"
click node_engine "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/pipeline.py"
click node_retriever "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/retrieval.py"
click node_answering "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/generation.py"
click node_passage_schema "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/schema.py"
click node_ingest "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/ingest.py"
click node_starter "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/data/sample/passages.jsonl"
click node_evaluation "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/evaluation.py"
click node_qa_data "https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/data/eval/qa.jsonl"

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_browser,node_cli,node_compare,node_explorer,node_upload toneBlue
class node_engine,node_retriever,node_answering,node_passage_schema toneAmber
class node_ingest,node_starter,node_extended toneMint
class node_evaluation,node_qa_data toneRose
class node_user,node_ollama toneIndigo
```

## Components

### User interfaces

- [Browser app](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/app.py): `app.py`
- [CLI commands](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/cli.py): `src/indicrag/cli.py`
- [Method comparison](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/app.py): `app.py`
- [Corpus explorer](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/app.py): `app.py`
- [Document upload](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/app.py): `app.py`

### Question answering

- [QA orchestration](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/pipeline.py): `src/indicrag/pipeline.py`
- [Passage retrieval](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/retrieval.py): `src/indicrag/retrieval.py`
- [Answer generation](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/generation.py): `src/indicrag/generation.py`
- [Passage schema](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/schema.py): `src/indicrag/schema.py`

### Corpus and ingestion

- [Document ingestion](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/ingest.py): `src/indicrag/ingest.py`
- [Starter passages](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/data/sample/passages.jsonl): `data/sample/passages.jsonl`
- **Expanded passages**

### Evaluation

- [Evaluation metrics](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/src/indicrag/evaluation.py): `src/indicrag/evaluation.py`
- [QA evaluation set](https://github.com/divyajyoti019/nlp-cross-lingual-project/blob/main/data/eval/qa.jsonl): `data/eval/qa.jsonl`

### Other components

- **Questioner**
- **Ollama service**

## Connections

- Questioner → Browser app: asks via (evidence: `README.md`)
- Questioner → CLI commands: asks via (evidence: `README.md`)
- Browser app → QA orchestration: submits questions (evidence: `README.md`)
- CLI commands → QA orchestration: runs ask (evidence: `src/indicrag/cli.py`)
- QA orchestration → Passage retrieval: searches passages (evidence: `src/indicrag/pipeline.py`)
- QA orchestration → Answer generation: generates answer (evidence: `src/indicrag/pipeline.py`)
- QA orchestration → Passage schema: uses passage data (evidence: `src/indicrag/pipeline.py`)
- Passage retrieval → Passage schema: ranks passages (evidence: `src/indicrag/retrieval.py`)
- Answer generation → Ollama service: calls when enabled (evidence: `src/indicrag/generation.py`)
- Browser app → Method comparison: offers comparison (evidence: `README.md`)
- Method comparison → Passage retrieval: compares methods (evidence: `README.md`)
- Browser app → Corpus explorer: offers search (evidence: `README.md`)
- Browser app → Document upload: offers uploads (evidence: `README.md`)
- Document upload → Document ingestion: ingests documents (evidence: `README.md`)
- CLI commands → Document ingestion: runs ingestion (evidence: `src/indicrag/cli.py`)
- Document ingestion → Passage schema: creates passages (evidence: `src/indicrag/ingest.py`)
- Starter passages → QA orchestration: loads corpus (evidence: `README.md`)
- Expanded passages → QA orchestration: loads optionally (evidence: `README.md`)
- CLI commands → Evaluation metrics: runs evaluation (evidence: `src/indicrag/cli.py`)
- Evaluation metrics → Passage retrieval: scores retrieval (evidence: `src/indicrag/evaluation.py`)
- Evaluation metrics → QA evaluation set: reads questions (evidence: `README.md`)
