ShuKe777/ensemble-cf-v2
AI model by ShuKe777
Older snapshot: check the original source for current information.
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
把一份家庭画像和一段月末市场摘要，答成三道投资决策题的 Laya checkpoint。编码器是 jhu-clsp/mmBERT-base（MIT，约 322M）的全量微调，上面是 Laya 0.4.2 的决策头。权重是 fp16。 这不是生成模型，也不能用 AutoModel.frompretrained 推理。加载入口是 laya.agent.Agent。 很多人时用 predictbatch，同一组 questions 打在一个 batch 上。maxlen 是 1024，headmaxlen 是 256。v2 精简 persona 在这个长度里不截断。 三道题合成一个人的目标持仓时，项目里的折算是： 期望权益金额是 资产 × P(持有) × E[比例｜持有]，再乘行业权重。资产档代表值是 0.5 万、3 万、12.5 万、35 万、100 万元。 state 是两段英文文本拼起来的： 20 个行业由申万一级合并而来，选项顺序就是 questions.json 里的顺序：农业、食品饮料、煤炭石油、钢铁有色、化工、建筑建材、房地产、机械、电力设备与新能源、国防、汽车、电子、计算机与通信、传媒、医药、银行、非银金融、家电纺服、零售服务、公用交通。 起点是 Laya multilingual（同一份 mmBERT-base）。scripts/finetunelaya.py 调用 laya.train.finetune，损失是软交叉熵 soft-ce，不是默认的 rlcd。编码器不冻结。每个 epoch 打乱 choice 题的选项顺序，避免学到位置先验。 四个 epoch 的训练损失是…
Read the full model card ↗ · Preview checked 2026-10-27T22:17:53.984Z
Inside the original model card — Document outline
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Documentation belongs to its respective authors. Reported project/model license: mit. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
ShuKe777/ensemble-cf-v2 is a task-unspecified model repository published by ShuKe777 on Hugging Face. A library was not reported.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- Not reported
- Library
- Not reported
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- 321,908,998
- Architecture
- Not reported
- License
- mit
- Access
- Not gated by Hugging Face
- Created
- 2026-10-15T06:21:25.000Z
- Last modified
- 2026-10-15T06:32:57.000Z
Compatibility and lineage
Reported languages: en
Reported base models: jhu-clsp/mmBERT-base
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-10-15T06:42:18.893Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.