Causal Discovery Toolbox: Uncover causal relationships in Python
About
This paper presents a new open source Python framework for causal discovery from observational data and domain background knowledge, aimed at causal graph and causal mechanism modeling. The 'cdt' package implements the end-to-end approach, recovering the direct dependencies (the skeleton of the causal graph) and the causal relationships between variables. It includes algorithms from the 'Bnlearn' and 'Pcalg' packages, together with algorithms for pairwise causal discovery such as ANM. 'cdt' is available under the MIT License at https://github.com/Diviyan-Kalainathan/CausalDiscoveryToolbox.
Diviyan Kalainathan, Olivier Goudet• 2019
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Causal Discovery | Synthetic DAGs | TPR0.83 | 125 | |
| DAG learning | Synthetic (test) | SID7.14e+3 | 101 | |
| Causal Discovery | Synthetic DAG data (test) | TPR89 | 40 | |
| Causal Discovery | Synthetic DAG data | TPR83 | 40 | |
| Causal Discovery | Synthetic Data | Runtime2.83e+3 | 21 | |
| Causal Discovery | Synthetic DAG Datasets | Runtime (s)3.73e+3 | 14 | |
| Learning Directed Acyclic Graphs | Synthetic DAGs Gumbel noise distribution 1.0 (test) | SHD350 | 5 | |
| Learning Directed Acyclic Graphs | Synthetic DAGs Default: 100 nodes, 400 edges, ER, Gaussian, n=1000 1.0 (test) | SHD325 | 5 | |
| Learning Directed Acyclic Graphs | Synthetic DAGs High edge density: 1000 edges 1.0 (test) | SHD961 | 4 | |
| Learning Directed Acyclic Graphs | Synthetic DAGs Dense root causes: p=0.5 1.0 (test) | SHD271 | 4 |
Showing 10 of 15 rows