algorithm/docs/api-examples.md

智能算法展示平台 API 使用示例

本文档提供了使用智能算法展示平台API的各种示例，涵盖从基础到高级的用法。

准备工作

在开始之前，请确保您已获得API访问令牌。

Python 示例

import requests
import json

# 设置基础URL和认证令牌
BASE_URL = "http://localhost:8001/api/v1"
headers = {
    "Authorization": "Bearer YOUR_ACCESS_TOKEN",
    "Content-Type": "application/json"
}

JavaScript 示例

const BASE_URL = "http://localhost:8001/api/v1";
const headers = {
    "Authorization": "Bearer YOUR_ACCESS_TOKEN",
    "Content-Type": "application/json"
};

算法管理示例

获取算法列表

Python

def get_algorithms():
    response = requests.get(f"{BASE_URL}/algorithms", headers=headers)
    if response.status_code == 200:
        return response.json()
    else:
        print(f"Error: {response.status_code} - {response.text}")
        return None

# 使用示例
algorithms = get_algorithms()
if algorithms:
    print(f"找到 {len(algorithms['algorithms'])} 个算法")
    for alg in algorithms['algorithms']:
        print(f"- {alg['name']} ({alg['type']})")

JavaScript

async function getAlgorithms() {
    try {
        const response = await fetch(`${BASE_URL}/algorithms`, {
            method: 'GET',
            headers: headers
        });
        
        if (response.ok) {
            const data = await response.json();
            console.log(`找到 ${data.algorithms.length} 个算法`);
            data.algorithms.forEach(alg => {
                console.log(`- ${alg.name} (${alg.type})`);
            });
            return data;
        } else {
            console.error(`Error: ${response.status} - ${await response.text()}`);
            return null;
        }
    } catch (error) {
        console.error('请求失败:', error);
        return null;
    }
}

获取特定算法详情

Python

def get_algorithm_details(algorithm_id):
    response = requests.get(f"{BASE_URL}/algorithms/{algorithm_id}", headers=headers)
    if response.status_code == 200:
        return response.json()
    else:
        print(f"Error: {response.status_code} - {response.text}")
        return None

# 使用示例
algorithm = get_algorithm_details("alg-001")
if algorithm:
    print(f"算法名称: {algorithm['name']}")
    print(f"算法类型: {algorithm['type']}")
    print(f"版本数量: {len(algorithm['versions'])}")

JavaScript

async function getAlgorithmDetails(algorithmId) {
    try {
        const response = await fetch(`${BASE_URL}/algorithms/${algorithmId}`, {
            method: 'GET',
            headers: headers
        });
        
        if (response.ok) {
            return await response.json();
        } else {
            console.error(`Error: ${response.status} - ${await response.text()}`);
            return null;
        }
    } catch (error) {
        console.error('请求失败:', error);
        return null;
    }
}

// 使用示例
getAlgorithmDetails("alg-001").then(algorithm => {
    if (algorithm) {
        console.log(`算法名称: ${algorithm.name}`);
        console.log(`算法类型: ${algorithm.type}`);
        console.log(`版本数量: ${algorithm.versions.length}`);
    }
});

算法调用示例

调用算法

Python

def call_algorithm(algorithm_id, version_id, input_data, params=None):
    payload = {
        "algorithm_id": algorithm_id,
        "version_id": version_id,
        "input_data": input_data
    }
    
    if params:
        payload["params"] = params
    
    response = requests.post(f"{BASE_URL}/algorithms/call", 
                            headers=headers, 
                            json=payload)
    
    if response.status_code == 200:
        return response.json()
    else:
        print(f"Error: {response.status_code} - {response.text}")
        return None

# 使用示例 - 图像分类算法
input_data = {
    "image_url": "https://example.com/sample-image.jpg"
}

params = {
    "confidence_threshold": 0.7
}

result = call_algorithm("alg-001", "ver-001", input_data, params)
if result:
    print(f"调用ID: {result['id']}")
    print(f"状态: {result['status']}")
    print(f"响应时间: {result['response_time']}秒")
    
    if result['status'] == 'success':
        predictions = result['output_data']['predictions']
        for pred in predictions:
            print(f"类别: {pred['class']}, 置信度: {pred['confidence']}")

JavaScript

async function callAlgorithm(algorithmId, versionId, inputData, params = null) {
    const payload = {
        algorithm_id: algorithmId,
        version_id: versionId,
        input_data: inputData
    };
    
    if (params) {
        payload.params = params;
    }
    
    try {
        const response = await fetch(`${BASE_URL}/algorithms/call`, {
            method: 'POST',
            headers: headers,
            body: JSON.stringify(payload)
        });
        
        if (response.ok) {
            return await response.json();
        } else {
            console.error(`Error: ${response.status} - ${await response.text()}`);
            return null;
        }
    } catch (error) {
        console.error('请求失败:', error);
        return null;
    }
}

// 使用示例
const inputData = {
    image_url: "https://example.com/sample-image.jpg"
};

const params = {
    confidence_threshold: 0.7
};

callAlgorithm("alg-001", "ver-001", inputData, params)
.then(result => {
    if (result) {
        console.log(`调用ID: ${result.id}`);
        console.log(`状态: ${result.status}`);
        console.log(`响应时间: ${result.response_time}秒`);
        
        if (result.status === 'success') {
            const predictions = result.output_data.predictions;
            predictions.forEach(pred => {
                console.log(`类别: ${pred.class}, 置信度: ${pred.confidence}`);
            });
        }
    }
});

获取调用结果

Python

def get_call_result(call_id):
    response = requests.get(f"{BASE_URL}/algorithms/calls/{call_id}", headers=headers)
    if response.status_code == 200:
        return response.json()
    else:
        print(f"Error: {response.status_code} - {response.text}")
        return None

# 使用示例
result = get_call_result("call-001")
if result:
    print(f"调用状态: {result['status']}")
    if result['status'] == 'success':
        print(f"输出数据: {json.dumps(result['output_data'], indent=2)}")

JavaScript

async function getCallResult(callId) {
    try {
        const response = await fetch(`${BASE_URL}/algorithms/calls/${callId}`, {
            method: 'GET',
            headers: headers
        });
        
        if (response.ok) {
            return await response.json();
        } else {
            console.error(`Error: ${response.status} - ${await response.text()}`);
            return null;
        }
    } catch (error) {
        console.error('请求失败:', error);
        return null;
    }
}

// 使用示例
getCallResult("call-001").then(result => {
    if (result) {
        console.log(`调用状态: ${result.status}`);
        if (result.status === 'success') {
            console.log(`输出数据:`, JSON.stringify(result.output_data, null, 2));
        }
    }
});

通过API网关调用算法

网关调用示例

Python

def call_algorithm_via_gateway(algorithm_id, version_id, input_data, params=None):
    payload = {
        "algorithm_id": algorithm_id,
        "version_id": version_id,
        "input_data": input_data
    }
    
    if params:
        payload["params"] = params
    
    response = requests.post(f"{BASE_URL}/gateway/call", 
                            headers=headers, 
                            json=payload)
    
    if response.status_code == 200:
        return response.json()
    else:
        print(f"Error: {response.status_code} - {response.text}")
        return None

# 使用示例
gateway_result = call_algorithm_via_gateway("alg-001", "ver-001", input_data, params)
if gateway_result:
    print(f"网关调用成功，响应时间: {gateway_result['response_time']}秒")

JavaScript

async function callAlgorithmViaGateway(algorithmId, versionId, inputData, params = null) {
    const payload = {
        algorithm_id: algorithmId,
        version_id: versionId,
        input_data: inputData
    };
    
    if (params) {
        payload.params = params;
    }
    
    try {
        const response = await fetch(`${BASE_URL}/gateway/call`, {
            method: 'POST',
            headers: headers,
            body: JSON.stringify(payload)
        });
        
        if (response.ok) {
            return await response.json();
        } else {
            console.error(`Error: ${response.status} - ${await response.text()}`);
            return null;
        }
    } catch (error) {
        console.error('请求失败:', error);
        return null;
    }
}

// 使用示例
callAlgorithmViaGateway("alg-001", "ver-001", inputData, params)
.then(gatewayResult => {
    if (gatewayResult) {
        console.log(`网关调用成功，响应时间: ${gatewayResult.response_time}秒`);
    }
});

历史记录示例

获取调用历史

Python

def get_call_history(skip=0, limit=100, algorithm_id=None, status=None):
    params = {"skip": skip, "limit": limit}
    if algorithm_id:
        params["algorithm_id"] = algorithm_id
    if status:
        params["status"] = status
    
    response = requests.get(f"{BASE_URL}/history/user-calls", 
                            headers=headers, 
                            params=params)
    
    if response.status_code == 200:
        return response.json()
    else:
        print(f"Error: {response.status_code} - {response.text}")
        return None

# 使用示例
history = get_call_history(limit=10, status="success")
if history:
    print(f"找到 {history['count']} 条历史记录")
    for record in history['history']:
        print(f"- {record['algorithm_id']}: {record['status']} ({record['response_time']}s)")

JavaScript

async function getCallHistory(skip = 0, limit = 100, algorithmId = null, status = null) {
    let url = `${BASE_URL}/history/user-calls?skip=${skip}&limit=${limit}`;
    if (algorithmId) url += `&algorithm_id=${algorithmId}`;
    if (status) url += `&status=${status}`;
    
    try {
        const response = await fetch(url, {
            method: 'GET',
            headers: headers
        });
        
        if (response.ok) {
            return await response.json();
        } else {
            console.error(`Error: ${response.status} - ${await response.text()}`);
            return null;
        }
    } catch (error) {
        console.error('请求失败:', error);
        return null;
    }
}

// 使用示例
getCallHistory(0, 10, null, "success").then(history => {
    if (history) {
        console.log(`找到 ${history.count} 条历史记录`);
        history.history.forEach(record => {
            console.log(`- ${record.algorithm_id}: ${record.status} (${record.response_time}s)`);
        });
    }
});

获取调用统计

Python

def get_call_statistics(user_id=None, algorithm_id=None):
    params = {}
    if user_id:
        params["user_id"] = user_id
    if algorithm_id:
        params["algorithm_id"] = algorithm_id
    
    response = requests.get(f"{BASE_URL}/history/statistics", 
                            headers=headers, 
                            params=params)
    
    if response.status_code == 200:
        return response.json()
    else:
        print(f"Error: {response.status_code} - {response.text}")
        return None

# 使用示例
stats = get_call_statistics()
if stats:
    print(f"总调用次数: {stats['total_calls']}")
    print(f"成功率: {stats['success_rate']}%")
    print(f"平均响应时间: {stats['avg_response_time']}秒")
    print(f"今日调用: {stats['today_calls']}")

JavaScript

async function getCallStatistics(userId = null, algorithmId = null) {
    let url = `${BASE_URL}/history/statistics`;
    const params = {};
    if (userId) params.user_id = userId;
    if (algorithmId) params.algorithm_id = algorithmId;
    
    if (Object.keys(params).length > 0) {
        const queryString = new URLSearchParams(params).toString();
        url += `?${queryString}`;
    }
    
    try {
        const response = await fetch(url, {
            method: 'GET',
            headers: headers
        });
        
        if (response.ok) {
            return await response.json();
        } else {
            console.error(`Error: ${response.status} - ${await response.text()}`);
            return null;
        }
    } catch (error) {
        console.error('请求失败:', error);
        return null;
    }
}

// 使用示例
getCallStatistics().then(stats => {
    if (stats) {
        console.log(`总调用次数: ${stats.total_calls}`);
        console.log(`成功率: ${stats.success_rate}%`);
        console.log(`平均响应时间: ${stats.avg_response_time}秒`);
        console.log(`今日调用: ${stats.today_calls}`);
    }
});

监控API示例

获取系统健康状况

Python

def get_health_status():
    response = requests.get(f"{BASE_URL}/monitoring/health", headers=headers)
    if response.status_code == 200:
        return response.json()
    else:
        print(f"Error: {response.status_code} - {response.text}")
        return None

# 使用示例
health = get_health_status()
if health:
    print(f"系统状态: {health['status']}")
    print(f"CPU使用率: {health['system_metrics']['cpu_percent']}%")
    print(f"内存使用率: {health['system_metrics']['memory_percent']}%")

JavaScript

async function getHealthStatus() {
    try {
        const response = await fetch(`${BASE_URL}/monitoring/health`, {
            method: 'GET',
            headers: headers
        });
        
        if (response.ok) {
            return await response.json();
        } else {
            console.error(`Error: ${response.status} - ${await response.text()}`);
            return null;
        }
    } catch (error) {
        console.error('请求失败:', error);
        return null;
    }
}

// 使用示例
getHealthStatus().then(health => {
    if (health) {
        console.log(`系统状态: ${health.status}`);
        console.log(`CPU使用率: ${health.system_metrics.cpu_percent}%`);
        console.log(`内存使用率: ${health.system_metrics.memory_percent}%`);
    }
});

获取算法性能指标

Python

def get_algorithm_performance(algorithm_id, days=7):
    response = requests.get(
        f"{BASE_URL}/monitoring/performance/algorithm/{algorithm_id}",
        headers=headers,
        params={"days": days}
    )
    if response.status_code == 200:
        return response.json()
    else:
        print(f"Error: {response.status_code} - {response.text}")
        return None

# 使用示例
performance = get_algorithm_performance("alg-001", 30)
if performance:
    print(f"算法 {performance['algorithm_id']} 的30天性能指标:")
    print(f"总调用次数: {performance['total_calls']}")
    print(f"成功率: {performance['success_rate']}%")
    print(f"平均响应时间: {performance['average_response_time']}秒")

JavaScript

async function getAlgorithmPerformance(algorithmId, days = 7) {
    try {
        const response = await fetch(
            `${BASE_URL}/monitoring/performance/algorithm/${algorithmId}?days=${days}`,
            {
                method: 'GET',
                headers: headers
            }
        );
        
        if (response.ok) {
            return await response.json();
        } else {
            console.error(`Error: ${response.status} - ${await response.text()}`);
            return null;
        }
    } catch (error) {
        console.error('请求失败:', error);
        return null;
    }
}

// 使用示例
getAlgorithmPerformance("alg-001", 30).then(performance => {
    if (performance) {
        console.log(`算法 ${performance.algorithm_id} 的30天性能指标:`);
        console.log(`总调用次数: ${performance.total_calls}`);
        console.log(`成功率: ${performance.success_rate}%`);
        console.log(`平均响应时间: ${performance.average_response_time}秒`);
    }
});

OpenAI集成示例

生成仿真数据

Python

def generate_simulation_data(prompt, data_type="text"):
    payload = {
        "prompt": prompt,
        "data_type": data_type
    }
    
    response = requests.post(f"{BASE_URL}/openai/generate-data", 
                            headers=headers, 
                            json=payload)
    
    if response.status_code == 200:
        return response.json()
    else:
        print(f"Error: {response.status_code} - {response.text}")
        return None

# 使用示例
simulation_data = generate_simulation_data("一张包含猫和狗的照片", "image")
if simulation_data:
    print(f"生成的仿真数据类型: {simulation_data['type']}")
    print(f"生成的数据: {simulation_data['data'][:100]}...")

JavaScript

async function generateSimulationData(prompt, dataType = "text") {
    const payload = {
        prompt: prompt,
        data_type: dataType
    };
    
    try {
        const response = await fetch(`${BASE_URL}/openai/generate-data`, {
            method: 'POST',
            headers: headers,
            body: JSON.stringify(payload)
        });
        
        if (response.ok) {
            return await response.json();
        } else {
            console.error(`Error: ${response.status} - ${await response.text()}`);
            return null;
        }
    } catch (error) {
        console.error('请求失败:', error);
        return null;
    }
}

// 使用示例
generateSimulationData("一张包含猫和狗的照片", "image")
.then(simulationData => {
    if (simulationData) {
        console.log(`生成的仿真数据类型: ${simulationData.type}`);
        console.log(`生成的数据: ${simulationData.data.substring(0, 100)}...`);
    }
});

错误处理

API调用可能会返回各种错误，以下是常见的错误处理方式：

Python

def safe_api_call(url, method="GET", payload=None):
    try:
        if method.upper() == "GET":
            response = requests.get(url, headers=headers)
        elif method.upper() == "POST":
            response = requests.post(url, headers=headers, json=payload)
        elif method.upper() == "PUT":
            response = requests.put(url, headers=headers, json=payload)
        elif method.upper() == "DELETE":
            response = requests.delete(url, headers=headers)
        
        # 检查HTTP状态码
        if response.status_code >= 200 and response.status_code < 300:
            return {"success": True, "data": response.json()}
        else:
            return {
                "success": False, 
                "error": f"HTTP {response.status_code}", 
                "message": response.text
            }
    
    except requests.exceptions.RequestException as e:
        return {"success": False, "error": "Request Error", "message": str(e)}
    except Exception as e:
        return {"success": False, "error": "Unknown Error", "message": str(e)}

# 使用示例
result = safe_api_call(f"{BASE_URL}/algorithms", "GET")
if result["success"]:
    print("API调用成功:", result["data"])
else:
    print("API调用失败:", result["error"], "-", result["message"])

JavaScript

async function safeApiCall(url, method = "GET", payload = null) {
    try {
        const options = {
            method: method.toUpperCase(),
            headers: headers
        };
        
        if (payload && (method.toUpperCase() === "POST" || method.toUpperCase() === "PUT")) {
            options.body = JSON.stringify(payload);
        }
        
        const response = await fetch(url, options);
        
        // 检查HTTP状态码
        if (response.ok) {
            return {
                success: true,
                data: await response.json()
            };
        } else {
            return {
                success: false,
                error: `HTTP ${response.status}`,
                message: await response.text()
            };
        }
    } catch (error) {
        return {
            success: false,
            error: "Request Error",
            message: error.message
        };
    }
}

// 使用示例
safeApiCall(`${BASE_URL}/algorithms`, "GET")
.then(result => {
    if (result.success) {
        console.log("API调用成功:", result.data);
    } else {
        console.log("API调用失败:", result.error, "-", result.message);
    }
});

高级用法

批量调用算法

Python

import asyncio
import aiohttp

async def batch_call_algorithms(calls_data):
    async with aiohttp.ClientSession() as session:
        tasks = []
        for call_data in calls_data:
            task = asyncio.create_task(
                async_call_algorithm(session, call_data)
            )
            tasks.append(task)
        
        results = await asyncio.gather(*tasks)
        return results

async def async_call_algorithm(session, call_data):
    headers = {
        "Authorization": "Bearer YOUR_ACCESS_TOKEN",
        "Content-Type": "application/json"
    }
    
    async with session.post(
        f"{BASE_URL}/algorithms/call",
        headers=headers,
        json=call_data
    ) as response:
        return await response.json()

# 使用示例
calls_data = [
    {
        "algorithm_id": "alg-001",
        "version_id": "ver-001",
        "input_data": {"image_url": "https://example.com/img1.jpg"}
    },
    {
        "algorithm_id": "alg-001",
        "version_id": "ver-001",
        "input_data": {"image_url": "https://example.com/img2.jpg"}
    }
]

# 注意：这需要在异步环境中运行
# results = asyncio.run(batch_call_algorithms(calls_data))

使用SDK进行API调用

我们还提供了官方SDK来简化API调用：

Python SDK示例

from algorithm_showcase.client import AlgorithmShowcaseClient

# 初始化客户端
client = AlgorithmShowcaseClient(
    base_url="http://localhost:8001",
    api_key="YOUR_API_KEY"
)

# 调用算法
try:
    result = client.call_algorithm(
        algorithm_id="alg-001",
        version_id="ver-001",
        input_data={"image_url": "https://example.com/sample.jpg"},
        params={"confidence_threshold": 0.7}
    )
    print(f"调用成功: {result}")
except Exception as e:
    print(f"调用失败: {e}")

JavaScript SDK示例

import { AlgorithmShowcaseClient } from '@algorithm-showcase/sdk';

// 初始化客户端
const client = new AlgorithmShowcaseClient({
    baseUrl: 'http://localhost:8001',
    apiKey: 'YOUR_API_KEY'
});

// 调用算法
try {
    const result = await client.callAlgorithm({
        algorithmId: 'alg-001',
        versionId: 'ver-001',
        inputData: { imageUrl: 'https://example.com/sample.jpg' },
        params: { confidenceThreshold: 0.7 }
    });
    console.log('调用成功:', result);
} catch (error) {
    console.error('调用失败:', error);
}

以上是智能算法展示平台API的完整使用示例。这些示例涵盖了从基本操作到高级用法的各个方面，可以帮助开发者快速上手并有效利用平台的功能。