隐私计算-4-SecretFlow

日向远 2024-03-23 23:54:14

SecretFlow安装

运行要求

安装方式

win11下，我选择利用pycharm下进wsl（ubuntu）新建anaconda，选择python3.8。环境内安装，我个人选择pypi方式。因为是新环境，先换pip源再下载。

pip3 config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
pip install -U secretflow
pip install -U secretflow-lite

仿真模式-单机仿真

创建个test1.py文件，运行

import secretflow as sf

sf.init(parties=['alice', 'bob'], address='local')
alice = sf.PYU('alice')
bob = sf.PYU('bob')
print(alice(lambda x : x + 1)(2))
print(bob(lambda x : x + 1)(2))

仿真模式-集群仿真

模拟alice参与方结点

# alice
ray start --head \
--node-ip-address="127.0.0.1" --port="8080" \
--resources='{"alice": 16} ' \
--include-dashboard=False \
--disable-usage-stats

模拟bob参与方结点

# bob
ray start \
--address="172.26.219.43:8080" \
--resources=' {"bob": 16} ' \
--disable-usage-stats

查看ray状态

初始化secretflow

创建个test2.py文件，运行

# test2.py
import secretflow as sf

sf.init(parties=['alice', 'bob'], address='127.0.0.1:8080')
alice = sf.PYU('alice')
bob = sf.PYU('bob')
print(alice(lambda x: x + 1)(2))
print(bob(lambda x: x + 1)(2))

创建SPU

创建个test3.py文件，运行

# test3.py
import spu
import secretflow as sf

sf.init(parties=['alice', 'bob'], address='127.0.0.1:8080')

cluster_def = {
    'nodes': [
        {
            'party': 'alice',
            'address': '127.0.0.1:8000',
            'listen_addr': '0.0.0.0:8000'
        },
        {
            'party': 'bob',
            'address': '127.0.0.1:8088',
            'listen_addr': '0.0.0.0:8088'
        },
    ],
    'runtime_config': {
        'protocol': spu.spu_pb2.SEMI2K,
        'field': spu.spu_pb2.FM128,
        'sigmoid_mode': spu.spu_pb2.RuntimeConfig.SIGMOID_REAL,
    }
}

spu = sf.SPU(cluster_def=cluster_def)