10 Commits

Author	SHA1	Message	Date
楚毓	c0d2c008b0	fix: bound LLM inference memory for batch screening	2026-09-23 10:11:41 +08:00
Michelle0574	a812694500	feat(app): 新增批量筛选，结果与模式可跨刷新恢复	2026-09-05 17:15:37 +00:00
Michelle0574	9810854ac7	fix: 修正bug	2026-08-16 13:51:46 +00:00
Michelle0574	48b22cfe86	fix: 收窄优化搜索域，新增 /predict 端点	2026-08-15 12:23:54 +00:00
Michelle0574	71f915e683	fix: 修正 size 预测的反归一化并将 MoE+LLM 模型接入应用	2026-08-14 11:24:02 +00:00
Michelle0574	866941039f	feat: final_train_optuna_cv 支持 MoE + Qwen2.5-7B QLoRA + soft-RAG，产出可部署 checkpoint	2026-08-13 10:05:00 +00:00
RYDE-WORK	28b181e194	数据预处理时自动存档总荧光强度的均值与方差,以便于预测时反演荧光强度	2026-03-03 15:48:59 +08:00
RYDE-WORK	9e4800b77e	内外部数据LNP四组分比例统一为百分比数,可电离脂质与mRNA比例的搜索范围与四组分解耦	2026-02-28 11:52:55 +08:00
RYDE-WORK	a9392aa780	Update models and UI	2026-02-11 16:49:28 +08:00
RYDE-WORK	982e98cced	Add LNP fomular optimization	2026-01-25 23:54:20 +08:00