AstrAI/astrai
ViperEkura d08a92c7bd feat: add frequency penalty to inference sampling pipeline
- Add FrequencyPenaltyStrategy (logit -= penalty * count)
- Per-task rep_window for penalty history lookup
- Wire through engine, task, executor, API layer
- Add --frequency_penalty and --rep_window to stream_chat.py
- 9 unit tests for frequency penalty strategy
2026-07-17 21:28:31 +08:00
..
config fix : SFT pipeline position_ids default & doc boundary preservation 2026-07-04 15:59:11 +08:00
dataset fix: rewrite GRPO data pipeline for offline record-level access 2026-07-17 14:34:41 +08:00
extension refactor: stride-based attn interface with layout and causal mask 2026-07-14 21:34:42 +08:00
inference feat: add frequency penalty to inference sampling pipeline 2026-07-17 21:28:31 +08:00
model perf: replace paged KV cache with contiguous ContiguousCache, decode all groups 2026-07-05 11:34:36 +08:00
parallel feat : add distributed checkpoint via executor checkpoint_context 2026-07-13 12:27:09 +08:00
preprocessing fix: rewrite GRPO data pipeline for offline record-level access 2026-07-17 14:34:41 +08:00
serialization feat: add --resume flag to decouple weight loading from training resumption 2026-07-16 14:23:23 +08:00
tokenize refactor : Pipeline 去除去重,ids 重命名为 sequence,泛型透传 2026-05-31 15:14:27 +08:00
trainer feat: add --resume flag to decouple weight loading from training resumption 2026-07-16 14:23:23 +08:00
__init__.py chore: bump version to 1.3.9 2026-07-12 21:04:20 +08:00
factory.py fix: ForwardRef._evaluate Python 3.12 compatibility 2026-07-03 18:41:19 +08:00
protocols.py refactor: 重构训练后端为 Executor 模式 2026-05-24 20:35:44 +08:00