绝大多数同学学习qlib,是从qlib提供的基本案例workflow_by_code.ipynb开始的,从文件名workfow可以看出,它隐含了一个基本的使用机器学习执行多因子预测的基本流(工作流)。 这个流程概括出来包括如下步骤: 设置股池, 如all或csi300 加载行情 生成因子和标签 因子和标签预处理 设置时段(训练、验证、测试) 训练 预测 ...
workflow_by_code.ipynb workflow_by_code.py qlib scripts tests .deepsource.toml .gitignore .mypy.ini .pre-commit-config.yaml .pylintrc .readthedocs.yml CHANGES.rst CODE_OF_CONDUCT.md LICENSE MANIFEST.in README.md SECURITY.md setup.pyBreadcrumbs qlib /examples/...
qrun /Users/liuhuanshuo/Downloads/qlib-main/examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158.yaml 启动瞬间,风扇就开始转了 经过一段时间等待后,终端会打印出执行过程中的关键参数与一些日志,除了命令行输出的这些信息之外,更多的过程信息会被写入主目录下的workflow_by_code.ipynb,我们启动Jupyter ...
🐛 Bug Description To Reproduce the workflow_by_code.py I am using pycharm remote debugging on the server and triggering OSError. Steps to reproduce the behavior: import qlib from qlib.constant import REG_CN from qlib.utils import init_in...
workflow_by_code可以看做是qlib的Hello World。 Qlib支持使用ymal配置。 port_analysis_config = { "executor": { "class": "SimulatorExecutor", "module_path": "qlib.backtest.executor", "kwargs": { "time_per_step":"day", "generate_portfolio_metrics": True, ...
使用Qilb内置的qrun运行上面的workflow 参考examples中workflow_by_code脚本,在Notebook中执行 参考run_all_model.py,写一个Python脚本,从命令行中使用python run_all_model.py --models=lightgbm执行 ” 运行多个模型 在Qlib中,我么可以使用run_all_model.py运行多个模型进行多次迭代,不过当前只能在Linux下执行,并...
参考examples中workflow_by_code脚本,在Notebook中执行 参考run_all_model.py,写一个Python脚本,从命令行中使用python run_all_model.py --models=lightgbm执行 ” 运行多个模型 在Qlib中,我么可以使用run_all_model.py运行多个模型进行多次迭代,不过当前只能在Linux下执行,并且该脚本将为每个模型创建一个唯一的虚拟...
# 开始实验with R.start(experiment_name="workflow"): #模型训练 R.log_params(**flatten_dict(CSI300_GBDT_TASK))model.fit(dataset) R.save_ozzbjects(**{"params.pkl": model})# 预测recorder = R.get_recorder() sr = SignalRecord(model, dataset, recorder) sr.generate()# 信号...
Tips: If you fail to installQlibor run the examples in your environment, comparing your steps and theCI workflowmay help you find the problem. Data Preparation Load and prepare data by running the following code: Get with module # get 1d datapython -m qlib.run.get_data qlib_data --targe...
Building Customized Quant Research Workflow by Code Quant Model Zoo Run a single model Run multiple models Quant Dataset Zoo More About Qlib Offline Mode and Online Mode Performance of Qlib Data Server Contributing Qlib is an AI-oriented quantitative investment platform, which aims to realize the po...