Files
CompanionGuard-RL/.gitignore
zhangsiyuan 766b4811be feat: port wangyu data pipeline and scripts into code/ structure
- code/src/data/: data_generator, dataset, llm_judge, __init__
  (multi-turn LLM dialogue generator, JSONL loader, LLM auto-annotator)
- code/scripts/: generate_siliconflow.py (SiliconFlow async generator, 701 lines)
  run_detector.sh / run_intervention.sh / run_full_pipeline.sh (launch scripts)
- code/configs/intervention_config.yaml: add reward.w1-w5 reference block
  (NOTE: v5 reward.py uses hardcoded constants; these fields are reference-only)
- .gitignore: fix data/ pattern to /data/ to avoid matching code/src/data/

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-18 14:59:48 +08:00

65 lines
808 B
Plaintext

# === 数据集(体积过大)===
/data/
code/data/
# === 模型权重 ===
code/checkpoints/
**/*.pt
**/*.bin
**/*.safetensors
**/*.h5
# === Python ===
**/__pycache__/
**/*.py[cod]
**/*.egg-info/
**/.pytest_cache/
dist/
build/
.eggs/
# === 虚拟环境 ===
**/.venv*/
**/venv/
**/env/
# === 历史压缩包 ===
tmp/archives/
sync_v*.tar.gz
sync_v*.zip
# === 大型数据 / 实验日志 ===
experiments/*.log
**/*.jsonl
**/*.json.gz
wandb/
# === 旧方向归档 ===
旧方向信息/
# === 工具配置(用户本地)===
.claude/
.playwright-mcp/
# === OS / 编辑器 ===
.DS_Store
Thumbs.db
.idea/
.vscode/
*.swp
# === 密钥 ===
.env
*.env
# === LaTeX 编译产物 ===
paper/*.aux
paper/*.bbl
paper/*.blg
paper/*.log
paper/*.toc
paper/*.out
paper/*.fls
paper/*.fdb_latexmk
paper/*.synctex.gz