Upload README.md with huggingface_hub

README.md

@@ -1,18 +1,68 @@
----
-language: ja
-license: cc-by-4.0
-task_categories:
-  - question-answering
-  - text-generation
-tags:
-  - data-engineering
-  - japanese
-  - instruction-tuning
-  - pretraining
-pretty_name: データエンジニア2026 学習データセット（日本語）
-size_categories:
-  - 1K<n<10K
----
-# データエンジニア2026 学習データセット（日本語）
-
-日本語のデータエンジニア向け学習データセット。SFT（Q&A）と事前学習コーパスを含みます。
+# Data Engineer 2026 (Japanese) Dataset
+
+A Japanese-language dataset for training and fine-tuning models on modern data engineering topics as of 2026. This dataset includes instruction-tuning (SFT) Q&A pairs and a pretraining corpus covering distributed processing, streaming, data lakehouse, orchestration, transformation, local analytics, modeling, quality/governance, cloud/data platform, MLOps, and AI/agentic data workflows.
+
+## Dataset splits
+
+| Split | Size | Description |
+|---|---|---|
+| `sft` | 1780 samples | Instruction-following Q&A pairs in Japanese (instruction/input/output). Covers definitions, comparisons, how-tos, design/best practices, debugging, interview-style, trends (2026), outlines, and aspect-specific questions. |
+| `pretrain` | 186 samples | Japanese technical explanatory text (chunked articles/notes) about data engineering topics for continued pretraining. |
+
+## Schema
+
+### SFT (sft)
+- `id` (str): sample ID
+- `category` (str): high-level category (分散処理エンジン/ストリーミング基盤/DWH・SQL/テーブル格式/オーケストレーション/変換/ローカル分析/モデリング/データ運用/クラウド基盤/AI・ML/現場/トレンド/移行/設計 etc.)
+- `topic` (str): topic slug (e.g. `spark`, `flink`, `kafka`, `iceberg`, `airflow`, `dbt`, `duckdb`, `polars`, `lakehouse`, `contracts`, `governance-ai`, ...)
+- `qtype` (str): question type (`faq`, `definition`, `overview`, `design`, `howto`, `debug`, `trend`, `interview`, `compare`, `outline`, `aspect`, `variant`)
+- `aspect` (str): aspect (`general` or `全体像/設計/パフォーマンス/コスト/運用/セキュリティ/移行/2026年動向`)
+- `difficulty` (str): `intermediate` (default)
+- `instruction` (str): question/instruction in Japanese
+- `input` (str): optional context (empty in this release)
+- `output` (str): answer/explanation in Japanese (markdown)
+- `lang` (str): `ja`
+- `source` (str): `curated+synthetic` or `synthetic`
+
+### Pretrain (pretrain)
+- `id` (str): sample ID
+- `text` (str): explanatory Japanese text (markdown-style)
+- `topic` (str): topic slug
+- `category` (str): category
+- `lang` (str): `ja`
+- `source` (str): `synthetic`
+
+## Coverage (key topics)
+
+- Distributed engines: Apache Spark 4.x, Apache Flink 2.x
+- Streaming: Apache Kafka 4.x (KRaft), streaming fundamentals, CDC (Debezium), streaming DBs (RisingWave/Materialize)
+- Lakehouse/table formats: Apache Iceberg, Delta Lake, Hudi, catalogs (Unity Catalog/Polaris/Gravitino/Nessie), formats (Parquet/Arrow/Avro)
+- Warehouses/SQL engines: BigQuery, Snowflake, Redshift, Trino, StarRocks, ClickHouse, SQL craft
+- Orchestration/transform: Apache Airflow 3.x (Task SDK), Dagster (SDA), dbt (Fusion/Core v2.0), SQLMesh
+- Local/modern analytics: DuckDB 1.5/2.0, Polars, Python data stack 2026 (pandas 3.x, uv, ruff, marimo, pydantic)
+- Data modeling/contracts: Kimball/Data Vault/OBT, data contracts (ODCS), data quality (Soda/Great Expectations/Deequ/Elementary), lineage (OpenLineage)
+- Cloud/platform: AWS/GCP/Azure data platforms, Kubernetes data (Spark/Flink operators/Strimzi/Kueue)
+- AI/ML/data for AI: AI/LLM data pipelines (embeddings/RAG/vector DB), MLOps data (feature store/lineage/versioning), governance for AI agents (MCP/text2sql)
+- Ops/FinOps/reliability, testing/CI, migration, design patterns
+
+All content reflects 2026-era trends (e.g. Iceberg as open table format, Spark Connect, Flink disaggregated state, Airflow 3, dbt Fusion, DuckDB/Wasm/Iceberg, streaming-first lakehouse, data contracts, agentic data workflows).
+
+## How to load
+
+```python
+from datasets import load_dataset
+
+ds = load_dataset("LoNebula/data-engineer-2026-ja")
+print(ds)
+# ds['sft'], ds['pretrain']
+```
+
+## Notes & License
+
+- This dataset is primarily synthetically generated from a curated Japanese knowledge base. It is intended for research/educational purposes and model training/evaluation.
+- While care was taken for factual accuracy (aligning to 2026 versions/trends), it may contain minor inaccuracies. Always verify against official documentation.
+- License: CC BY 4.0
+
+## Acknowledgments
+
+Generated with assistance from Kilo/Claude Code tooling. Dataset structure and schema are designed for instruction-tuning (SFT) and continued pretraining in Japanese.