190 lines
7.3 KiB
Python
190 lines
7.3 KiB
Python
|
"""
|
|||
|
|
分析路由
|
|||
|
|
"""
|
|||
|
|
|
|||
|
|
import logging
|
|||
|
|
import json
|
|||
|
|
from datetime import datetime
|
|||
|
|
from typing import Optional, Dict, Any, List
|
|||
|
|
|
|||
|
|
from fastapi import APIRouter, HTTPException, status, BackgroundTasks
|
|||
|
|
from pydantic import BaseModel
|
|||
|
|
import psutil
|
|||
|
|
import os
|
|||
|
|
import gc
|
|||
|
|
import shutil
|
|||
|
|
|
|||
|
|
from app.core.config import settings
|
|||
|
|
from app.services.analysis import TimeSeriesAnalysisSystem
|
|||
|
|
|
|||
|
|
logger = logging.getLogger(__name__)
|
|||
|
|
router = APIRouter()
|
|||
|
|
|
|||
|
|
|
|||
|
|
class AnalysisRequest(BaseModel):
|
|||
|
|
"""分析请求模型"""
|
|||
|
|
filename: str
|
|||
|
|
file_type: str = "csv"
|
|||
|
|
task_description: str = "时间序列数据分析"
|
|||
|
|
data_background: Dict[str, Any] = {}
|
|||
|
|
original_image: Optional[str] = None
|
|||
|
|
language: str = "zh"
|
|||
|
|
generate_plots: bool = False
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
@router.get("/available_methods", summary="获取可用的分析方法")
|
|||
|
|
async def get_available_methods() -> dict:
|
|||
|
|
"""获取所有可用的分析方法"""
|
|||
|
|
return {
|
|||
|
|
"success": True,
|
|||
|
|
"methods": {
|
|||
|
|
'statistical_overview': {'name': '统计概览', 'description': '生成数据的基本统计信息和分布图表'},
|
|||
|
|
'time_series_analysis': {'name': '时间序列分析', 'description': '分析变量随时间变化的趋势和模式'},
|
|||
|
|
'acf_pacf_analysis': {'name': '自相关分析', 'description': '生成自相关和偏自相关函数图'},
|
|||
|
|
'stationarity_tests': {'name': '平稳性检验', 'description': '执行ADF、KPSS等平稳性检验'},
|
|||
|
|
'normality_tests': {'name': '正态性检验', 'description': '执行Shapiro-Wilk、Jarque-Bera正态性检验'},
|
|||
|
|
'seasonal_decomposition': {'name': '季节性分解', 'description': '分解时间序列的趋势、季节和残差成分'},
|
|||
|
|
'spectral_analysis': {'name': '频谱分析', 'description': '分析时间序列的频域特征'},
|
|||
|
|
'correlation_analysis': {'name': '相关性分析', 'description': '计算变量间的相关性并生成热力图'},
|
|||
|
|
'pca_scree_plot': {'name': 'PCA碎石图', 'description': '显示主成分分析的解释方差'},
|
|||
|
|
'pca_analysis': {'name': '主成分分析', 'description': '降维分析，识别数据的主要变化方向'},
|
|||
|
|
'feature_importance': {'name': '特征重要性', 'description': '分析各变量对目标预测的重要性'},
|
|||
|
|
'clustering_analysis': {'name': '聚类分析', 'description': '将数据点分组为具有相似特征的簇'},
|
|||
|
|
'factor_analysis': {'name': '因子分析', 'description': '识别潜在的因子结构'},
|
|||
|
|
'cointegration_test': {'name': '协整检验', 'description': '检验时间序列变量间的长期均衡关系'},
|
|||
|
|
'var_analysis': {'name': '向量自回归', 'description': '多变量时间序列建模和预测'}
|
|||
|
|
}
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
def check_memory():
|
|||
|
|
"""检查内存使用"""
|
|||
|
|
process = psutil.Process(os.getpid())
|
|||
|
|
memory_mb = process.memory_info().rss / 1024 / 1024
|
|||
|
|
logger.info(f"当前内存使用: {memory_mb:.2f} MB")
|
|||
|
|
|
|||
|
|
if memory_mb > settings.MAX_MEMORY_MB:
|
|||
|
|
logger.warning(f"内存使用超过阈值 ({settings.MAX_MEMORY_MB} MB)，执行垃圾回收")
|
|||
|
|
gc.collect()
|
|||
|
|
|
|||
|
|
|
|||
|
|
@router.post("/analyze", summary="执行完整分析")
|
|||
|
|
async def analyze_data(request: AnalysisRequest, background_tasks: BackgroundTasks) -> dict:
|
|||
|
|
"""
|
|||
|
|
执行完整的时间序列分析
|
|||
|
|
|
|||
|
|
流程:
|
|||
|
|
1. 加载并预处理数据
|
|||
|
|
2. 执行15种分析方法
|
|||
|
|
3. 调用AI API 进行深度分析
|
|||
|
|
4. 生成PDF/PPT/HTML报告
|
|||
|
|
"""
|
|||
|
|
try:
|
|||
|
|
logger.info("=" * 60)
|
|||
|
|
logger.info(f"开始分析: {request.filename}")
|
|||
|
|
logger.info(f"任务: {request.task_description}")
|
|||
|
|
logger.info(f"语言: {request.language}")
|
|||
|
|
logger.info("=" * 60)
|
|||
|
|
|
|||
|
|
# 检查内存
|
|||
|
|
check_memory()
|
|||
|
|
|
|||
|
|
# 检查文件存在
|
|||
|
|
file_path = settings.get_upload_path(request.filename)
|
|||
|
|
if not file_path.exists():
|
|||
|
|
raise HTTPException(
|
|||
|
|
status_code=status.HTTP_404_NOT_FOUND,
|
|||
|
|
detail=f"文件未找到: {request.filename}"
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
# 语言处理：支持 zh/en，其他值回退为 zh
|
|||
|
|
lang_key = request.language if request.language in {"zh", "en"} else "zh"
|
|||
|
|
|
|||
|
|
# charts 模式下强制不生成图片，即使请求传了 generate_plots=true
|
|||
|
|
generate_plots = False
|
|||
|
|
if request.generate_plots:
|
|||
|
|
logger.info("generate_plots requested true, forcing false to skip image generation")
|
|||
|
|
|
|||
|
|
# 创建分析器实例
|
|||
|
|
logger.info(f"初始化分析器 ({lang_key})...")
|
|||
|
|
analyzer = TimeSeriesAnalysisSystem(
|
|||
|
|
str(file_path),
|
|||
|
|
request.task_description,
|
|||
|
|
data_background=request.data_background,
|
|||
|
|
language=lang_key,
|
|||
|
|
generate_plots=generate_plots
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
# 运行分析
|
|||
|
|
logger.info("执行分析...")
|
|||
|
|
results_zh, log_zh = analyzer.run_analysis()
|
|||
|
|
|
|||
|
|
if results_zh is None:
|
|||
|
|
logger.error("中文分析失败")
|
|||
|
|
raise HTTPException(
|
|||
|
|
status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
|
|||
|
|
detail="分析失败"
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
logger.info("中文分析完成")
|
|||
|
|
|
|||
|
|
# 准备返回数据
|
|||
|
|
response_data = {
|
|||
|
|
"success": True,
|
|||
|
|
"meta": {
|
|||
|
|
"filename": request.filename,
|
|||
|
|
"task_description": request.task_description,
|
|||
|
|
"language": lang_key,
|
|||
|
|
"generate_plots": generate_plots,
|
|||
|
|
"created_at": datetime.now().isoformat(),
|
|||
|
|
},
|
|||
|
|
"analysis": {
|
|||
|
|
lang_key: {
|
|||
|
|
"pdf_filename": None,
|
|||
|
|
"ppt_filename": None,
|
|||
|
|
"data_description": results_zh.get("data_description"),
|
|||
|
|
"preprocessing_steps": results_zh.get("preprocessing_steps", []),
|
|||
|
|
"api_analysis": results_zh.get("api_analysis", {}),
|
|||
|
|
"steps": results_zh.get("steps", []),
|
|||
|
|
"charts": results_zh.get("charts", {}),
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
"images": {},
|
|||
|
|
"log": log_zh[-20:] if log_zh else [],
|
|||
|
|
"original_image": request.original_image if request.file_type == 'image' else None,
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
# 兼容旧前端：始终提供 analysis.zh
|
|||
|
|
if lang_key != "zh":
|
|||
|
|
response_data["analysis"]["zh"] = response_data["analysis"][lang_key]
|
|||
|
|
|
|||
|
|
analysis_bucket = response_data["analysis"][lang_key]
|
|||
|
|
|
|||
|
|
# 去除任何遗留的 image_path（兼容旧结构）
|
|||
|
|
steps = analysis_bucket.get("steps")
|
|||
|
|
if isinstance(steps, list):
|
|||
|
|
for step in steps:
|
|||
|
|
if isinstance(step, dict) and "image_path" in step:
|
|||
|
|
step.pop("image_path", None)
|
|||
|
|
|
|||
|
|
# images 保持为空兼容旧前端
|
|||
|
|
response_data["images"] = {}
|
|||
|
|
|
|||
|
|
logger.info("分析完成")
|
|||
|
|
return response_data
|
|||
|
|
|
|||
|
|
except HTTPException:
|
|||
|
|
raise
|
|||
|
|
except Exception as e:
|
|||
|
|
logger.error(f"分析异常: {str(e)}", exc_info=True)
|
|||
|
|
raise HTTPException(
|
|||
|
|
status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
|
|||
|
|
detail=str(e)
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
|
|||
|
|
|