replysadiq/qatar-customs-nl2sql
Qatar Customs NL2SQL — Arabic Natural Language to SAP HANA SQL Fine-tuning dataset for converting Arabic natural language questions into SAP HANA SQL queries for the Qatar General Directorate of Customs (GDC) database. Dataset Overview Metric Value Total question sets 99 Training examples 320 (3 fuzz levels × 80 questions + 80 follow-ups) Test examples 76 (3 fuzz levels × 19 questions + 19 follow-ups) Language Arabic 🇶🇦 (formal + dialectal)… See the full description on the dataset page: https://huggingface.co/datasets/replysadiq/qatar-customs-nl2sql.
Revert to bf16 — L4 GPU supports native bf16
Fix: use fp16 instead of bf16 for T4 GPU (Colab)
Add comprehensive README
Add dataset build script
Add evaluation script
Add training script (Qwen2.5-Coder-7B QLoRA)
Move schema.sql to eval/
Add test metadata to eval/ directory (not auto-loaded)
Remove test_with_meta.jsonl (conflicts with auto-loader schema)
Add HANA schema DDL
Add test data with metadata (question_num, fuzz_level, is_followup)
Add test data (76 examples)
Add training data (320 examples, Arabic NL → HANA SQL)
initial commit