update (#39)

Co-authored-by: hong.wang <[email protected]>

.gitignore

@@ -106,19 +106,27 @@ tmp/
 .DS_Store
 
 # dataset
+/dataset
 *.csv
 *.npz
-*.npy
-/dataset
+*.meta
+*.html
+.*
 
 # model
 *.pt
 checkpoints/
 .vscode/
+.ssh/
+.vscode-server/
+.ray/
+.cache/
+.bash_history
+.gitconfig
 cache/.DS_Store
 logs/
 examples/naive_horizontal_resnet/data/
 python/tree_structure
 python/tree_structure.png
-_book/
-node_modules/
+docs/zh_CN/build/
+docs/en/build/

README.md

@@ -1,6 +1,8 @@
+### [中文](./README_zh.md)
+
 [![License](https://img.shields.io/github/license/paritybit-ai/XFL)](https://opensource.org/licenses/Apache-2.0)
 [![Documentation Status](https://readthedocs.org/projects/xfl/badge/?version=latest)](https://xfl.readthedocs.io/en/latest/?badge=latest)
-
+[![coverage report](https://git.basebit.me/bag1/federated-learning/badges/master/coverage.svg)](https://git.basebit.me/bag1/federated-learning/-/commits/master)
 
 XFL is a high-performance, high-flexibility, high-applicability, lightweight, open and easy-to-use Federated Learning framework.
 It supports a variety of federation models in both horizontal and vertical federation scenarios. 
@@ -13,14 +15,14 @@ and applies secure communication protocols to ensure communication security.
   - High-performance algorithm library
 
     - Comprehensive algorithms: support a variety of mainstream horizontal/vertical federation algorithms.
-    - Excellent performance: significantly exceeds the average performance of federated learning products. 
+    - Excellent performance: significantly exceeds the average performace of federated learning products. 
     - Network optimization: adapt to high latency, frequent packet loss, and unstable network environments.
 
   - Flexible deployment
 
-    - Parties: support two-party/multi-party federated learning.
-    - Schedulering: any participant can act as a task scheduler.
-    - Hardware: support CPU/GPU/hybrid deployment.
+    - parties: support two-party/multi-party federated learning.
+    - schedulering: any participant can act as a task scheduler.
+    - hardware: support CPU/GPU/hybrid deployment.
 
   - Lightweight, open and easy to use:
 
@@ -65,9 +67,10 @@ sh run.sh
 ```
 
 - [Quick Start](./docs/en/source/tutorial/usage.md)
-# Document
 
+# Document
 - [Document](https://xfl.readthedocs.io/en/latest)
+
 ## Tutorial
 - [Introduction](./docs/en/source/tutorial/introduction.md)
 

README_zh.md

@@ -0,0 +1,45 @@
+### [English](./README.md) 
+
+[![License](https://img.shields.io/github/license/paritybit-ai/XFL)](https://opensource.org/licenses/Apache-2.0)
+[![Documentation Status](https://readthedocs.org/projects/xfl/badge/?version=latest)](https://xfl.readthedocs.io/en/latest/?badge=latest)
+[![coverage report](https://git.basebit.me/bag1/federated-learning/badges/master/coverage.svg)](https://git.basebit.me/bag1/federated-learning/-/commits/master)
+
+XFL是一个高性能、高灵活度、高适用性、轻量开放的联邦学习框架，支持横向联邦、纵向联邦等多种联邦模型，运用同态加密、差分隐私、多方安全计算等多种加密计算技术保护用户的原始数据不被泄露，使用安全通信协议保护通信安全，使用户在合法合规的基础上进行联合建模，实现数据价值。
+
+# 项目特点
+  - 高性能的算法库
+
+    - 算法全面：支持多种主流横向/纵向联邦算法
+    - 性能优异：性能大幅领先于信通院性能评测平均水平
+    - 网络优化：在弱网络、高延迟、大量丢包、较长时间断网的情况下依然可以完成训练任务
+
+  - 应用部署灵活
+
+    - 计算节点灵活： 支持两方/多方计算节点部署
+    - 算力调配灵活： 无论有无label，支持任何一方为发起方，支持辅助计算节点部署在任一方
+    - 安装部署灵活： 支持CPU/GPU/混合部署
+
+  - 轻量开放
+
+    - 轻量： 对服务器性能要求低，部分算法可在性能较差的环境下运行
+    - 开放：支持 Pytorch / Tensorflow 等主流机器学习框架，支持用户自定义横向模型
+
+# [快速开始](./docs/zh_CN/source/tutorial/usage.md)
+
+# Document
+- [Document](https://xfl.readthedocs.io/en/latest)
+
+## Tutorial
+- [简介](./docs/en/source/tutorial/introduction.md)
+
+## Algorithms
+- [算法列表](./docs/en/source/algorithms/algorithms_list.rst)
+- [密码算法](./docs/en/source/algorithms/cryptographic_algorithm.rst)
+- [差分隐私](./docs/en/source/algorithms/differential_privacy.rst)
+
+## Development
+- [API](./docs/en/source/development/api.rst)
+- [开发指南](./docs/en/source/development/algos_dev.rst)
+
+# License
+[Apache License 2.0](./LICENSE)

demo/horizontal/binning_woe_iv/2party/config/fed_conf.json → demo/horizontal/gcn_mol/2party/config/fed_conf.json

@@ -1,6 +1,6 @@
 {
     "fed_info": {
-       "scheduler": {
+        "scheduler": {
             "scheduler": "localhost:55001"
         },
         "trainer": {
@@ -14,5 +14,5 @@
     "redis_server": "localhost:6379",
     "grpc": {
         "use_tls": false
-    } 
+    }
 }

demo/horizontal/poisson_regression/2party_env/config/train_config_assit_trainer.json → demo/horizontal/gcn_mol/2party/config/trainer_config_assist_trainer.json

@@ -2,18 +2,24 @@
     {
         "identity": "assist_trainer",
         "model_info": {
-            "name": "horizontal_poisson_regression",
+            "name": "horizontal_gcn_mol",
             "config": {
-                "input_dim": 3,
-                "bias": true
+                "input_dim": 74,
+                "num_gnn_layers": 1,
+                "gnn_hidden_feats": 256,
+                "activation": "relu",
+                "dropout": 0.001,
+                "predictor_hidden_feats": 512,
+                "batchnorm": false,
+                "residual": false
             }
         },
         "input": {
             "valset": [
                 {
                     "type": "csv",
-                    "path": "./dataset/arizona_card_hos",
-                    "name": "ari_card_hos_test.csv",
+                    "path": "./dataset/hiv/2party",
+                    "name": "hiv_test.csv",
                     "has_label": true,
                     "has_id": false
                 }
@@ -24,7 +30,7 @@
             "model": {
                 "type": "file",
                 "path": "/opt/checkpoints/[JOB_ID]/[NODE_ID]",
-                "name": "horizontal_poisson_regression.pt"
+                "name": "horizontal_resnet.pt"
             },
             "metrics": {
                 "type": "csv",
@@ -40,17 +46,17 @@
         "train_info": {
             "device": "cpu",
             "interaction_params": {
-                "save_frequency": -1,
+                "save_frequency": 2,
                 "save_probabilities": true,
                 "save_probabilities_bins_number": 10,
                 "write_training_prediction": true,
                 "write_validation_prediction": true,
                 "echo_training_metrics": true
             },
             "params": {
-                "global_epoch": 30,
+                "global_epoch": 50,
                 "local_epoch": 1,
-                "batch_size": 32,
+                "batch_size": 128,
                 "aggregation_config": {
                     "type": "fedavg",
                     "encryption": {
@@ -68,12 +74,16 @@
                     }
                 },
                 "lossfunc_config": {
-                    "PoissonNLLLoss": {
-                        "log_input": false
-                    }
+                    "BCEWithLogitsLoss": {}
                 },
                 "metric_config": {
-                    "mean_poisson_deviance": {}
+                    "accuracy": {},
+                    "roc_auc_score": {}
+                },
+                "early_stopping": {
+                    "key": "acc",
+                    "patience": 2,
+                    "delta": 0
                 },
                 "extra_config": {}
             }