dowhy
causalnex
dowhy | causalnex | |
---|---|---|
8 | 2 | |
6,781 | 2,151 | |
1.7% | 1.3% | |
8.8 | 5.4 | |
2 days ago | 5 days ago | |
Python | Python | |
MIT License | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
dowhy
-
Causality for Machine Learning
I'm a fan of the Do Why library out of Microsoft. Even as a novice in the field of causal modeling it can get you up and running by estimating the causal graph based on your data. https://github.com/py-why/dowhy
-
Acceptable data formats for Predictive Stepwise Logistic Regression
considering how well understood the generating process is, causal analysis could potentially be very powerful here and would model the "not every possible combination of variables is represented" component extremely naturally. https://github.com/py-why/dowhy
-
Do you use any specific framework when it comes to causal inference?
The Do-why package could be useful.
-
Causal Explanations Considered Harmful: On the logical fallacy of causal projection
Here's one from Microsoft! https://github.com/py-why/dowhy
-
[Q] What are some of the most useful topics/classes in philosophy for Statistics?
Before those discussions, it's good to understand the very basics of the topic so you 1) demonstrate momentum to the prof, and 2) have the basis for a meaningful discussion. For causal reasoning, you can check out the Pearl book Causal inference in statistics, a primer, which is short and readable. Definitely check out the Do Why python package which has good tutorials and videos.
- [R] DoWhy-GCM: An extension of DoWhy for causal inference in graphical causal models
- DoWhy is a Python library for causal inference
causalnex
-
How many of you still buy and read textbooks after your degree?
I don't claim to defend that this is actually the right way of dealing with those things, but QuantumBlack gave a talk at neurips a couple years back, and really hyped up their package (https://github.com/quantumblacklabs/causalnex) for dealing with this stuff.
-
What are some tools/best practices that Causal Inferencing teams use for experimentation?
As for causal libraries I'd recommend CausalNex, it's the only library that I know that does Judea Pearl's do() operator, and I think that's really great if you want to intervene over causal knowledge (that you'll want).
What are some alternatives?
looper - A resource list for causality in statistics, data science and physics
causalml - Uplift modeling and causal inference with machine learning algorithms
causal-learn - Causal Discovery in Python. It also includes (conditional) independence tests and score functions.
pgmpy - Python Library for learning (Structure and Parameter), inference (Probabilistic and Causal), and simulations in Bayesian Networks.
causalgraph - A python package for modeling, persisting and visualizing causal graphs embedded in knowledge graphs.
scikit-learn - scikit-learn: machine learning in Python
CausalPy - A Python package for causal inference in quasi-experimental settings
causaldag - Python package for the creation, manipulation, and learning of Causal DAGs
Causality
pyphi - A toolbox for integrated information theory.
Keras - Deep Learning for humans