# ai-admin-algorithm

**Repository Path**: captainwilson/ai-admin-algorithm

## Basic Information

- **Project Name**: ai-admin-algorithm
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-01-02
- **Last Updated**: 2026-01-02

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Financial Data Analysis API

This service provides FastAPI endpoints for analyzing financial journal Excel files.

## Installation

```bash
pip install -r requirements.txt
```

## Running the Service

```bash
python run.py
```

The API will be available at `http://localhost:6113`

## API Endpoints

### POST /analyze_journal

Analyzes an uploaded Excel file containing financial journal data.

**Request:**
- Method: POST
- Content-Type: multipart/form-data
- Body: file (Excel file: .xlsx, .xls, or .csv)

**Response:**
```json
{
  "success": true,
  "data": {
    "expense_by_category": [
      {"name": "Category1", "value": 1000.0},
      {"name": "Category2", "value": 500.0}
    ],
    "income_by_category": [
      {"name": "Category1", "value": 2000.0},
      {"name": "Category2", "value": 1500.0}
    ],
    "timeline_data": {
      "dates": ["2024-01-01", "2024-01-02"],
      "expenses": [100.0, 200.0],
      "incomes": [500.0, 600.0],
      "capitals": [0.0, 0.0]
    }
  }
}
```

### GET /health

Health check endpoint.

**Response:**
```json
{
  "status": "ok"
}
```

## Excel File Format

The service automatically detects columns in the Excel file. It looks for:

- **Date column**: Contains dates (columns named: date, 日期, 时间, time)
- **Amount column**: Contains numeric amounts (columns named: amount, 金额, 收入, 支出)
- **Category column**: Contains category descriptions (columns named: category, 类别, 分类, 项目, description, 描述, 摘要)
- **Type column** (optional): Contains transaction type (expense/income/capital, 支出/收入/资本)

If type column is not found, the service infers the type from:
1. The amount sign (negative = expense, positive = income)
2. Category name keywords (expense/income/capital related keywords)

## Notes

- The service supports .xlsx, .xls, and .csv file formats
- Date parsing is flexible and handles various date formats
- Amounts are normalized (expenses are always positive values)
- Categories are grouped and summed for visualization