AcidMap/Cd_Prediction_Integrated_System/main.py

"""
主执行脚本 - Cd预测集成系统
Main execution script for Cd Prediction Integrated System

Description: 集成运行作物Cd模型、有效态Cd模型和数据分析可视化的完整流程
架构简化版本：直接调用通用绘图模块，删除中间封装层
"""

import os
import sys
import logging
from datetime import datetime

# 添加项目根目录到Python路径
sys.path.append(os.path.dirname(os.path.abspath(__file__)))
# 添加上级目录到路径以访问通用绘图模块
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))

import config
from models.crop_cd_model.predict import CropCdPredictor
from models.effective_cd_model.predict import EffectiveCdPredictor
from analysis.data_processing import DataProcessor
from utils.common import setup_logging, validate_files

# 直接导入通用绘图模块
from app.utils.mapping_utils import csv_to_raster_workflow, MappingUtils

def main():
    """
    主执行函数
    执行完整的Cd预测分析流程
    """
    # 设置日志
    log_file = os.path.join(config.OUTPUT_PATHS["reports_dir"], 
                           f"execution_log_{datetime.now().strftime('%Y%m%d_%H%M%S')}.log")
    setup_logging(log_file)
    logger = logging.getLogger(__name__)
    
    logger.info("=" * 60)
    logger.info("开始执行Cd预测集成系统 (架构简化版本)")
    logger.info("=" * 60)
    
    try:
        # 确保目录存在
        config.ensure_directories()
        logger.info("项目目录结构检查完成")
        
        # 获取当前工作流配置（支持运行时动态配置）
        workflow_config = config.get_workflow_config()
        logger.info(f"当前工作流配置: {workflow_config}")
        
        # 步骤1: 运行作物Cd模型预测
        if workflow_config["run_crop_model"]:
            logger.info("步骤1: 运行作物Cd模型预测...")
            crop_predictor = CropCdPredictor()
            crop_output = crop_predictor.predict()
            logger.info(f"作物Cd模型预测完成，输出文件: {crop_output}")
        
        # 步骤2: 运行有效态Cd模型预测  
        if workflow_config["run_effective_model"]:
            logger.info("步骤2: 运行有效态Cd模型预测...")
            effective_predictor = EffectiveCdPredictor()
            effective_output = effective_predictor.predict()
            logger.info(f"有效态Cd模型预测完成，输出文件: {effective_output}")
        
        # 步骤3: 数据整合处理
        if workflow_config["combine_predictions"]:
            logger.info("步骤3: 整合预测结果与坐标数据...")
            data_processor = DataProcessor()
            final_data_files = data_processor.combine_predictions_with_coordinates()
            logger.info(f"数据整合完成，生成文件: {list(final_data_files.keys())}")
        
        # 初始化通用绘图工具（用于可视化）
        mapping_utils = MappingUtils()
        
        # 步骤4: 为每个模型生成栅格文件和可视化图
        model_outputs = {}
        
        for model_name, final_data_file in final_data_files.items():
            if model_name == 'combined':  # 跳过合并文件的可视化
                continue
            
            # 只为当前工作流配置中启用的模型生成输出文件
            if model_name == 'crop_cd' and not workflow_config.get("run_crop_model", False):
                logger.info(f"跳过{model_name}模型的可视化（工作流配置中未启用）")
                continue
            if model_name == 'effective_cd' and not workflow_config.get("run_effective_model", False):
                logger.info(f"跳过{model_name}模型的可视化（工作流配置中未启用）")
                continue
                
            logger.info(f"为{model_name}模型生成栅格和可视化...")
            
            # 为每个模型确定显示名称
            if model_name == 'crop_cd':
                display_name = "作物Cd"
                title_name = "Crop Cd Prediction"
            elif model_name == 'effective_cd':
                display_name = "有效态Cd"
                title_name = "Effective Cd Prediction"
            else:
                display_name = model_name
                title_name = f"{model_name} Prediction"
            
            model_outputs[model_name] = {}
            
            # 步骤4a: 生成栅格文件（直接调用通用绘图模块）
            if workflow_config["generate_raster"]:
                logger.info(f"步骤4a: 为{display_name}模型生成栅格文件...")
                
                # 使用csv_to_raster_workflow直接生成栅格
                try:
                    # 获取栅格配置（支持运行时参数覆盖）
                    raster_config = config.get_raster_config()
                    
                    workflow_result = csv_to_raster_workflow(
                        csv_file=final_data_file,
                        template_tif=config.ANALYSIS_CONFIG["template_tif"],
                        output_dir=config.OUTPUT_PATHS["raster_dir"],
                        boundary_shp=config.ANALYSIS_CONFIG.get("boundary_shp"),
                        resolution_factor=raster_config["resolution_factor"],
                        interpolation_method=raster_config["interpolation_method"],
                        field_name=raster_config["field_name"],
                        lon_col=raster_config["coordinate_columns"]["longitude"],
                        lat_col=raster_config["coordinate_columns"]["latitude"], 
                        value_col=raster_config["coordinate_columns"]["value"],
                        enable_interpolation=raster_config["enable_interpolation"]
                    )
                    
                    output_raster = workflow_result['raster']
                    
                    # 重命名为特定模型的文件
                    model_raster_path = os.path.join(
                        config.OUTPUT_PATHS["raster_dir"],
                        f"output_{model_name}.tif"
                    )
                    
                    if os.path.exists(output_raster) and output_raster != model_raster_path:
                        import shutil
                        shutil.move(output_raster, model_raster_path)
                        output_raster = model_raster_path
                    
                    model_outputs[model_name]['raster'] = output_raster
                    logger.info(f"{display_name}模型栅格文件生成完成: {output_raster}")
                    
                    # 清理中间shapefile文件
                    try:
                        shapefile_path = workflow_result.get('shapefile')
                        if shapefile_path and os.path.exists(shapefile_path):
                            # 删除shapefile及其相关文件
                            import glob
                            base_path = os.path.splitext(shapefile_path)[0]
                            for ext in ['.shp', '.shx', '.dbf', '.prj', '.cpg']:
                                file_to_delete = base_path + ext
                                if os.path.exists(file_to_delete):
                                    os.remove(file_to_delete)
                                    logger.debug(f"已删除中间文件: {file_to_delete}")
                            logger.info(f"已清理中间shapefile文件: {shapefile_path}")
                    except Exception as cleanup_error:
                        logger.warning(f"清理中间文件失败: {str(cleanup_error)}")
                    
                except Exception as e:
                    logger.error(f"栅格生成失败: {str(e)}")
                    raise
            
            # 步骤4b: 创建可视化地图（直接调用通用绘图模块）
            if workflow_config["create_visualization"]:
                logger.info(f"步骤4b: 为{display_name}模型创建可视化地图...")
                
                # 为每个模型创建独立的地图输出路径
                map_output_path = os.path.join(
                    config.OUTPUT_PATHS["figures_dir"],
                    f"Prediction_results_{model_name}"
                )
                
                try:
                    # 直接调用通用绘图模块的create_raster_map
                    map_output = mapping_utils.create_raster_map(
                        shp_path=config.ANALYSIS_CONFIG.get("boundary_shp"),
                        tif_path=output_raster,
                        output_path=map_output_path,
                        title=title_name,
                        colormap=config.VISUALIZATION_CONFIG["color_maps"][config.VISUALIZATION_CONFIG["default_colormap"]],
                        output_size=config.VISUALIZATION_CONFIG["figure_size"],
                        dpi=config.VISUALIZATION_CONFIG["dpi"],
                        resolution_factor=1.0,  # 保持栅格原始分辨率
                        enable_interpolation=False,  # 不在可视化阶段插值
                        interpolation_method='nearest'
                    )
                    
                    model_outputs[model_name]['map'] = map_output
                    logger.info(f"{display_name}模型地图可视化完成: {map_output}")
                    
                except Exception as e:
                    logger.error(f"地图可视化失败: {str(e)}")
                    raise
            
            # 步骤4c: 创建直方图（直接调用通用绘图模块）
            if workflow_config["create_histogram"]:
                logger.info(f"步骤4c: 为{display_name}模型创建预测值分布直方图...")
                
                # 为每个模型创建独立的直方图输出路径
                histogram_output_path = os.path.join(
                    config.OUTPUT_PATHS["figures_dir"],
                    f"Prediction_frequency_{model_name}.jpg"
                )
                
                # 为了避免中文字体问题，使用英文标题
                if model_name == 'crop_cd':
                    hist_title = 'Crop Cd Prediction Frequency'
                    hist_xlabel = 'Crop Cd Content (mg/kg)'
                elif model_name == 'effective_cd':
                    hist_title = 'Effective Cd Prediction Frequency'
                    hist_xlabel = 'Effective Cd Content (mg/kg)'
                else:
                    hist_title = f'{model_name} Prediction Frequency'
                    hist_xlabel = f'{model_name} Content'
                
                try:
                    # 直接调用通用绘图模块的create_histogram
                    histogram_output = mapping_utils.create_histogram(
                        file_path=output_raster,
                        save_path=histogram_output_path,
                        figsize=(6, 6),
                        xlabel=hist_xlabel,
                        ylabel='Frequency',
                        title=hist_title,
                        dpi=config.VISUALIZATION_CONFIG["dpi"]
                    )
                    
                    model_outputs[model_name]['histogram'] = histogram_output
                    logger.info(f"{display_name}模型直方图创建完成: {histogram_output}")
                    
                except Exception as e:
                    logger.error(f"直方图生成失败: {str(e)}")
                    raise
        
        logger.info("=" * 60)
        logger.info("所有流程执行完成！")
        logger.info("=" * 60)
        
        # 打印结果摘要
        print_summary(model_outputs)
        
    except Exception as e:
        logger.error(f"执行过程中发生错误: {str(e)}")
        logger.exception("详细错误信息:")
        raise
    
def print_summary(model_outputs):
    """打印执行结果摘要"""
    print("\n" + "=" * 60)
    print("[SUCCESS] Cd预测集成系统执行完成! (架构简化版本)")
    print("=" * 60)
    print(f"[INFO] 总输出目录: {config.OUTPUT_PATHS}")
    print()
    
    # 显示每个模型的输出文件
    for model_name, outputs in model_outputs.items():
        if model_name == 'crop_cd':
            display_name = "作物Cd模型"
        elif model_name == 'effective_cd':
            display_name = "有效态Cd模型"
        else:
            display_name = f"{model_name}模型"
            
        print(f"[MODEL] {display_name}:")
        for output_type, file_path in outputs.items():
            if output_type == 'raster':
                print(f"   [RASTER] 栅格文件: {file_path}")
            elif output_type == 'map':
                print(f"   [MAP] 地图文件: {file_path}")
            elif output_type == 'histogram':
                print(f"   [CHART] 直方图文件: {file_path}")
        print()
    
    print(f"[LOG] 日志文件目录: {config.OUTPUT_PATHS['reports_dir']}")
    print("=" * 60)

if __name__ == "__main__":
    main()