Duplicate from birdsql/bird23-train-filtered

Co-authored-by: Xiaolong Li <xia01ongLi@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,59 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.lz4 filter=lfs diff=lfs merge=lfs -text
+*.mds filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+# Audio files - uncompressed
+*.pcm filter=lfs diff=lfs merge=lfs -text
+*.sam filter=lfs diff=lfs merge=lfs -text
+*.raw filter=lfs diff=lfs merge=lfs -text
+# Audio files - compressed
+*.aac filter=lfs diff=lfs merge=lfs -text
+*.flac filter=lfs diff=lfs merge=lfs -text
+*.mp3 filter=lfs diff=lfs merge=lfs -text
+*.ogg filter=lfs diff=lfs merge=lfs -text
+*.wav filter=lfs diff=lfs merge=lfs -text
+# Image files - uncompressed
+*.bmp filter=lfs diff=lfs merge=lfs -text
+*.gif filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.tiff filter=lfs diff=lfs merge=lfs -text
+# Image files - compressed
+*.jpg filter=lfs diff=lfs merge=lfs -text
+*.jpeg filter=lfs diff=lfs merge=lfs -text
+*.webp filter=lfs diff=lfs merge=lfs -text
+# Video files - compressed
+*.mp4 filter=lfs diff=lfs merge=lfs -text
+*.webm filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,89 @@
+---
+license: cc-by-sa-4.0
+task_categories:
+- table-question-answering
+- question-answering
+language:
+- en
+size_categories:
+- 1K<n<10K
+---
+# BIRD-SQL Train (Filtered)
+
+A high-quality subset of the original BIRD train split for text-to-SQL finetuning.
+
+## Overview
+
+Over the past year the community has shared many observations about data quality in BIRD. We performed a rigorous data quality check process to retain examples that are **consistent with schema** and **faithfully answer the question**. The resulting set keeps **6,601** instances out of **9,428** (≈70%), and serves as a drop-in replacement for training.
+
+- **Original Train:** 9,428
+- **Filtered Train (this release):** 6,601
+
+The example code for training and inference can be found [here](https://github.com/bird-bench/mini_dev/tree/main)
+
+### For New Users
+If you are new to BIRD project, you can download the complete databases for the training set using the following link:
+[Download BIRD Train](https://bird-bench.oss-cn-beijing.aliyuncs.com/train.zip)
+
+### For Existing Users
+If you have already downloaded the BIRD training databases, you can pull the latest filtered data updates through Hugging Face using the following scripts:
+
+```python
+from datasets import load_dataset
+# Load the dataset
+dataset = load_dataset("birdsql/bird23-train-filtered")
+# Access the dataset
+print(dataset["train"][0])
+```
+
+You can find the column meaning json file [here](https://huggingface.co/datasets/birdsql/bird23-train-filtered/resolve/main/train_column_meaning.json) the key is composed of `database_id|table_name|column_name`, and the value is key information about each column and their value summarized from raw CSVs.
+
+## Training Quality
+We validate by finetuning a single open model with a standard SFT recipe and evaluating on the official BIRD Mini Dev and Dev set. We use the [Qwen/Qwen2.5-3B](https://huggingface.co/Qwen/Qwen2.5-3B) as the base model and follow the [Arctic-Text2SQL-R1 project ](https://www.snowflake.com/en/product/ai/ai-research/) data processing. You can find the original [repo](https://github.com/snowflakedb/ArcticTraining/tree/main/projects/arctic_text2sql_r1) and [paper](https://arxiv.org/abs/2505.20315) here. 
+
+### Performance Comparison (EX)
+
+| Setting                   | Mini-Dev | Dev      |
+| ------------------------- | -------- | -------- |
+| **Baseline**              | 26.2     | 31.88    |
+| **Original Train** | 45.4     | 50.46    |
+| **Filtered Train** | **46.0** | **50.0** |
+
+Takeaway: with **~30%** fewer training items, the filtered set matches the full set on the Mini-Dev and Dev set.
+
+### Data scaling on the filtered set
+
+
+<p align="left">
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/653693cb8ee17cfd44eed8ce/6_0KJzy4o1GfMDnqP3nA1.png" width="520">
+</p>
+
+## Dataset Introduction
+
+The dataset contains the main following resources:
+
+- `database`: The database should be stored under the [`./train_databases/`](./train_databases/). In each database folder, it has two components:
+  - `database_description`: the csv files are manufactured to describe database schema and its values for models to explore or references.
+  - `sqlite`: The database contents in BIRD.
+- `data`: Each text-to-SQL pairs with the oracle knowledge evidence is stored as a JSONL file, i.e., `train.jsonl`. It has four main parts:
+  - `db_id`: the names of databases
+  - `question`: the questions curated by human crowdsourcing according to database descriptions, database contents.
+  - `evidence`: the external knowledge evidence annotated by experts for assistance of models or SQL annotators.
+  - `SQL`: SQLs annotated by crowdsource referring to database descriptions, database contents, to answer the questions accurately.
+
+## Acknowledgements
+This work builds on the BIRD benchmark and the efforts of its creators and contributors. We thank the community for continuous feedback that helped shape this release.
+
+
+
+## Citation
+Please cite the repo if you think our work is helpful to you.
+```
+@article{li2024can,
+  title={Can llm already serve as a database interface? a big bench for large-scale database grounded text-to-sqls},
+  author={Li, Jinyang and Hui, Binyuan and Qu, Ge and Yang, Jiaxi and Li, Binhua and Li, Bowen and Wang, Bailin and Qin, Bowen and Geng, Ruiying and Huo, Nan and others},
+  journal={Advances in Neural Information Processing Systems},
+  volume={36},
+  year={2024}
+}
+```