恢复(正则图推断)
regain的Python项目详细描述
重新获得
考虑潜在变量影响的多时间戳正则化图推理。 它继承了scikit-learn包的功能。
入门
依赖性
REGAIN
需要:
- Python(>;=2.7或>;=3.5)
- 纽比(>;=1.8.2)
- scikit学习(>;=0.17)
您可以通过运行以下命令安装(必需的)依赖项:
pip install -r requirements.txt
要通过高斯过程优化使用参数选择,需要skopt。
安装
安装retain的最简单方法是使用pip
pip install regain
或conda
conda install -c fdtomasi regain
如果您想从源代码安装,或者想为项目做出贡献(例如,通过github发送pull请求),请继续阅读。在github中克隆存储库并将其添加到$pythonpath中。
git clone https://github.com/fdtomasi/regain.git
cd regain
python setup.py develop
快速启动
如何使用ltgl的简单示例。
importnumpyasnpfromregain.covarianceimportLatentTimeGraphLassofromregain.datasetsimportmake_datasetfromregain.utilsimporterror_norm_timenp.random.seed(42)data=make_dataset(n_dim_lat=1,n_dim_obs=10)X=data.datatheta=data.thetasmdl=LatentTimeGraphLasso(max_iter=50).fit(X)print("Error: %.2f"%error_norm_time(theta,mdl.precision_))
注意,LatentTimeGraphLasso
的输入是一个形状为(n_times, n_samples, n_dimensions)
的三维矩阵。
如果您只有一次(n_times = 1
),请确保在使用LatentTimeGraphLasso
之前使用X = X.reshape(1, *X.shape)
,或者,也可以使用LatentGraphLasso
。
引文
REGAIN
出现在以下两个出版物中。
对于LatentTimeGraphLasso
请使用
@inproceedings{Tomasi:2018:LVT:3219819.3220121, author = {Tomasi, Federico and Tozzo, Veronica and Salzo, Saverio and Verri, Alessandro}, title = {Latent Variable Time-varying Network Inference}, booktitle = {Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery \&\#38; Data Mining}, series = {KDD '18}, year = {2018}, isbn = {978-1-4503-5552-0}, location = {London, United Kingdom}, pages = {2338--2346}, numpages = {9}, url = {http://doi.acm.org/10.1145/3219819.3220121}, doi = {10.1145/3219819.3220121}, acmid = {3220121}, publisher = {ACM}, address = {New York, NY, USA}, keywords = {convex optimization, graphical models, latent variables, network inference, time-series}, }
以及用于TimeGraphLassoForwardBackward
plase
@InProceedings{pmlr-v72-tomasi18a, title = {Forward-Backward Splitting for Time-Varying Graphical Models}, author = {Tomasi, Federico and Tozzo, Veronica and Verri, Alessandro and Salzo, Saverio}, booktitle = {Proceedings of the Ninth International Conference on Probabilistic Graphical Models}, pages = {475--486}, year = {2018}, editor = {Kratochv\'{i}l, V\'{a}clav and Studen\'{y}, Milan}, volume = {72}, series = {Proceedings of Machine Learning Research}, address = {Prague, Czech Republic}, month = {11--14 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v72/tomasi18a/tomasi18a.pdf}, url = {http://proceedings.mlr.press/v72/tomasi18a.html}, abstract = {Gaussian graphical models have received much attention in the last years, due to their flexibility and expression power. However, the optimisation of such complex models suffer from computational issues both in terms of convergence rates and memory requirements. Here, we present a forward-backward splitting (FBS) procedure for Gaussian graphical modelling of multivariate time-series which relies on recent theoretical studies ensuring convergence under mild assumptions. Our experiments show that a FBS-based implementation achieves, with very fast convergence rates, optimal results with respect to ground truth and standard methods for dynamical network inference. Optimisation algorithms which are usually exploited for network inference suffer from drawbacks when considering large sets of unknowns. Particularly for increasing data sets and model complexity, we argue for the use of fast and theoretically sound optimisation algorithms to be significant to the graphical modelling community.}}