You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
73 lines
1.9 KiB
73 lines
1.9 KiB
# 开源项目贡献者流失预测
|
|
|
|
## 项目结构
|
|
'''
|
|
contributor_churn_prediction/
|
|
│
|
|
├── data/
|
|
│ ├── linux_commits.csv
|
|
│ └── rust_commits.csv
|
|
│
|
|
├── src/
|
|
│ ├── data_preprocessing.py
|
|
│ ├── time_series_prediction.py
|
|
│ ├── model.py
|
|
│
|
|
├── requirements.txt
|
|
├── main.py
|
|
└── README_V2.md
|
|
'''
|
|
|
|
## 环境配置
|
|
|
|
1. 安装 Anaconda 或 Miniconda。
|
|
|
|
2. 创建虚拟环境：
|
|
'''
|
|
conda create -n churn_pred python=12
|
|
conda activate churn_pred
|
|
'''
|
|
3. 安装依赖：
|
|
'''
|
|
pip install -r requirements.txt
|
|
'''
|
|
4. 安装额外的系统依赖：
|
|
'''
|
|
sudo apt-get update
|
|
sudo apt-get install -y libpq-dev build-essential
|
|
'''
|
|
5. 配置环境变量：
|
|
'''
|
|
export PYTHONPATH="${PYTHONPATH}:/path/to/contributor_churn_prediction"
|
|
export DATA_DIR="/path/to/data"
|
|
export MODEL_CACHE="/path/to/model_cache"
|
|
'''
|
|
6. 下载并安装自定义的模型包：
|
|
'''
|
|
git clone https://github.com/custom_models/churn_pred.git
|
|
cd churn_pred
|
|
pip install -e .
|
|
'''
|
|
## 运行代码
|
|
|
|
1. 数据预处理：
|
|
'''
|
|
python src/data_preprocessing.py --input $DATA_DIR/linux_commits.csv --output $DATA_DIR/processed_linux.pkl
|
|
python src/data_preprocessing.py --input $DATA_DIR/rust_commits.csv --output $DATA_DIR/processed_rust.pkl
|
|
'''
|
|
|
|
2. 特征工程：
|
|
'''
|
|
python src/time_series_prediction.py --input $DATA_DIR/processed_linux.pkl --output $DATA_DIR/features_linux.pkl
|
|
python src/time_series_prediction.py --input $DATA_DIR/processed_rust.pkl --output $DATA_DIR/features_rust.pkl
|
|
'''
|
|
3. 模型训练：
|
|
'''
|
|
python src/model.py --input $DATA_DIR/features_linux.pkl --model-type rf --output $MODEL_CACHE/model_linux.pkl
|
|
python src/model.py --input $DATA_DIR/features_rust.pkl --model-type xgb --output $MODEL_CACHE/model_rust.pkl
|
|
'''
|
|
4.预测：
|
|
'''
|
|
python main.py --linux-model $MODEL_CACHE/model_linux.pkl --rust-model $MODEL_CACHE/model_rust.pkl --output results.json
|
|
'''
|