Add some draft doc.

Author: JerryLife

README.md

@@ -0,0 +1,19 @@
+# VertiBench: Vertical Federated Learning Benchmark
+
+## Introduction
+
+VertiBench is a benchmark for [federated learning](https://ieeexplore.ieee.org/abstract/document/9599369/), [split learning](https://arxiv.org/abs/1912.12115), and [assisted learning](https://proceedings.neurips.cc/paper_files/paper/2022/hash/4d6938f94ab47d32128c239a4bfedae0-Abstract-Conference.html) on vertical partitioned data. It provides tools to synthetic vertical partitioned data from a given global dataset. VertiBench supports partition under various **imbalance** and **correlation** level, effectively simulating a wide-range of real-world vertical federated learning scenarios. 
+
+
+![data-dist-full.png](fig%2Fdata-dist-full.png)
+
+## Installation
+
+VertiBench has already been published on PyPI. The installation requires the installation of `python>=3.9`. To further install VertiBench, run the following command:
+
+```bash
+pip install vertibench
+```
+
+## Getting Started
+

example/EvaluatorExample.py

@@ -0,0 +1,9 @@
+from sklearn.datasets import make_classification
+
+# Generate a large dataset
+X, y = make_classification(n_samples=10000, n_features=10)
+
+
+from vertibench import Evaluator
+
+evaluator = Evaluator.ImportanceEvaluator(sample_rate=0.1)