oscarzhang/HealthRecommendation

Health Recommendation Model

一个基于 Qwen3-VL-4B-Instruct 的健康干预方案生成模型。

Model Description

这个模型是专门针对健康推荐和干预方案生成任务进行微调的 LoRA 适配器。它能够根据用户提供的症状信息，生成个性化的健康干预方案，包括生活方式建议、心理干预和环境改造建议。

主要功能

• 症状分析: 基于用户提供的症状信息进行分析
• 干预方案生成: 生成包含生活方式、心理、环境三个维度的干预建议
• 理论依据: 方案基于 P3 框架或 PROCEED 框架
• 评估指标: 提供可量化的评估指标

模型特点

• Base Model: Qwen/Qwen3-VL-4B-Instruct
• Fine-tuning Method: LoRA (Low-Rank Adaptation)
• Language: 中文 (Chinese)
• Domain: Healthcare, Health Recommendation

Model Details

Model Description

• Developed by: oscarzhang
• Model type: LoRA Adapter for Qwen3-VL-4B-Instruct
• Language(s): Chinese (中文)
• License: Apache 2.0 (inherited from base model)
• Finetuned from: Qwen/Qwen3-VL-4B-Instruct

Model Sources [optional]

• Repository: [More Information Needed]
• Paper [optional]: [More Information Needed]
• Demo [optional]: [More Information Needed]

Uses

Direct Use

这个模型可以直接用于生成健康干预方案。用户需要提供：

• 粗症状（主要症状描述）
• 细化症状（具体症状细节）

模型将输出包含以下内容的干预方案：

1. 推荐干预：
   • 生活方式：具体的行动建议
   • 心理：心理认知层面的干预
   • 环境：环境改造建议
2. 评估指标：可量化测量的指标

使用示例

from peft import PeftModel, PeftConfig
from transformers import AutoModelForCausalLM, AutoTokenizer

# 加载基础模型和适配器
base_model = AutoModelForCausalLM.from_pretrained(
    "Qwen/Qwen3-VL-4B-Instruct",
    trust_remote_code=True
)
model = PeftModel.from_pretrained(base_model, "oscarzhang/HealthRecommendation")
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-VL-4B-Instruct", trust_remote_code=True)

# 生成干预方案
prompt = """请基于症状信息，生成个性化的健康干预方案。

粗症状：疲劳
细化症状：早晨起床后持续疲惫感，影响日常工作"""
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=500, temperature=0.3)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

Downstream Use

这个模型可以进一步微调用于：

• 特定疾病的干预方案生成
• 不同年龄段人群的健康建议
• 结合更多健康数据的推荐系统

Out-of-Scope Use

⚠️ 重要限制：

• 本模型仅提供生活干预建议，不做医疗诊断
• 不应用于紧急医疗情况
• 对于需要医疗干预的情况，必须明确指出需要就医
• 不应替代专业医疗建议

Bias, Risks, and Limitations

[More Information Needed]

Recommendations

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

How to Get Started with the Model

Use the code below to get started with the model.

[More Information Needed]

Training Details

Training Data

训练数据包含126个健康干预方案样本，每个样本包含：

• 症状信息（粗症状和细化症状）
• 理论框架分析（P3框架和PROCEED框架）
• 完整的干预方案（生活方式、心理、环境三个维度）
• 评估指标

Training Procedure

Preprocessing

训练数据经过以下预处理：

• 症状标准化
• 理论框架提取
• 干预方案结构化
• 质量评分验证

Training Hyperparameters

• Training method: LoRA (Low-Rank Adaptation)
• Training regime: fp16 mixed precision
• LoRA rank (r): 16
• LoRA alpha: 16
• LoRA dropout: 0.05
• Target modules: v_proj, q_proj, k_proj, o_proj
• Learning rate: 2e-4 (with linear decay and warmup)
• Batch size: 1 (with gradient accumulation of 8)
• Training steps: 100
• Max sequence length: 512 tokens

Training Results

• Initial Loss: 2.5
• Final Loss: 0.7
• Loss reduction: 72%
• Training time: ~1.7 hours

Evaluation

Testing Data, Factors & Metrics

Testing Data

[More Information Needed]

Factors

[More Information Needed]

Metrics

[More Information Needed]

Results

[More Information Needed]

Summary

Model Examination [optional]

[More Information Needed]

Environmental Impact

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

• Hardware Type: [More Information Needed]
• Hours used: [More Information Needed]
• Cloud Provider: [More Information Needed]
• Compute Region: [More Information Needed]
• Carbon Emitted: [More Information Needed]

Technical Specifications [optional]

Model Architecture and Objective

[More Information Needed]

Compute Infrastructure

[More Information Needed]

Hardware

[More Information Needed]

Software

[More Information Needed]

Citation [optional]

BibTeX:

[More Information Needed]

APA:

[More Information Needed]

Glossary [optional]

[More Information Needed]

More Information [optional]

[More Information Needed]

Model Card Authors [optional]

[More Information Needed]

Model Card Contact

[More Information Needed]

Framework versions

• PEFT 0.17.1