One-stop time series analysis tool, supporting time series data preprocessing, feature engineering, model training, model evaluation, and model prediction.
Project description
PipelineTS
一站式时间序列分析工具，支持时序数据预处理、特征工程、模型训练、模型评估、模型预测等。
安装
conda install -c conda-forge prophet
python -m pip install PipelineTS
快速开始
from PipelineTS.dataset import LoadWebSales
init_data = LoadWebSales()[['date', 'type_a']]
valid_data = init_data.iloc[-30:, :]
data = init_data.iloc[:-30, :]
from PipelineTS.pipeline import PipelineTS
# list all models
PipelineTS.list_models()
from sklearn.metrics import mean_absolute_error
pipeline = PipelineTS(
time_col='date',
target_col='type_a',
lags=30,
random_state=42,
metric=mean_absolute_error,
metric_less_is_better=True
)
# training all models
pipeline.fit(data, valid_df=valid_data)
# use best model to predict next 30 steps data point
res = pipeline.predict(30)
数据准备
# TODO
预处理
# TODO
特征工程
# TODO
模型训练
# TODO
模型评估
# TODO
模型预测
# TODO
模型部署
# TODO
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
PipelineTS-0.2.2.tar.gz
(22.7 kB
view hashes)
Built Distribution
PipelineTS-0.2.2-py3-none-any.whl
(38.9 kB
view hashes)
Close
Hashes for PipelineTS-0.2.2-py3-none-any.whl
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | f7a00ccb7ab8d4cd5e3f8f2201f3a4d9656aed65be93c611f8391ff940f55b36 |
|
| MD5 | 9af7c288c7a8fe09281f7549cadcaa75 |
|
| BLAKE2b-256 | fcde17a308d7a9991a7f3dddfbdb10544792e39474c9e7405253149c62d78162 |
