Testimonials

• "At Quantopian we use PyMC3 to track uncertainty in the performance of a trading algorithm." - Thomas Wiecki
• "We use PyMC3 to evaluate A/B test performance. Works great with very little code!" - Thomas Hunger, We Are Wizards
• "PyMC3 is used at VoiceBox Technologies to compare algorithm performances using Kruschke's BEST algorithm. More is in development."
• "Used in research code at Channel 4 for developing internal forecasting tools." - Peader Coyle
• "At Managed by Q, we use PyMC3 for all of our statistical modeling, including A/B test analysis, sales forecasting, and churn prediction." - Daniel Weitzenfeld
• "PyMC3 is my primary tool for statistical modeling at Salesforce. I use it to combine disparate sources of information and pretty much anywhere that quantifying uncertainty is important. For example, we build hierarchical models to evaluate varying effects in web experiments and then to build meta-analyses that quantify the expected returns of a subsequent experiment. We've also been experimenting with gaussian processes to model time series data for forecasting." - Eddie Landesberg. Manager, Data Scientist
• "At Sounds we use PyMC3 in production to serve A/B testing results and for all our statistical modeling: computing lifetime values, predicting churn, defining user sessions, etc.. It is an invaluable tool; it contains cutting-edge algorithms, and yet its API is simple and intuitive enough to allow team members who are not familiar with Bayesian statistics to get up to speed really quickly." - Remi Louf
• "At Intercom, we've adopted PyMC3 as part of our A/B testing framework. The API made it easy to integrate into our existing experimentation framework and the methodology has made communication of experiment results much more straightforward for non technical stakeholders." - Louis Ryan
• "At the Novartis Institutes for Biomedical Research, we use PyMC3 for a wide variety of scientific and business use cases. The API is incredible, making it easy to express probabilistic models of our scientific experimental data and business processes, such as models of electrophysiology and molecular dose response." - Eric J. Ma
• "In the Integrated Fisheries Lab, we use PyMC3 to build statistical models that help us solve a wide-range of fisheries problems to improve conservation and management." - Aaron MacNeil
• "At The Black Tux we used PyMC3 to forecast reservations for future event weekends by building a probabilistic model in PyMC3 that reproduced previous business modeling efforts in Excel and added valuable uncertainty bounds to the forecast. It was easy to add more components, such as a second warehouse location, and the corresponding uncertainty to the forecast output. We also used PyMC3 to model the uncertainty in SKU/Size demand and future inventory pressure, as well as conversion curves for marketing cohorts by channel. The ability to code a probabilistic model in Python using an easy to use API that so closely represents the business' understanding of it was incredibly valuable!" -Claus Herther
• At SpaceX PyMC3 helped us estimate supplier delivery uncertainty and quantify which suppliers were consistent in sending us rocket parts and which suppliers we needed to partner with to understand how we could reduce variability - Ravin Kumar
• "We use PyMC3 in our cognitive modelling work. We use it for both parameter estimation and model comparison, alongside experimental data, to help us better understand human decision making." - Benjamin T. Vincent
• "I use PyMC3 to discover latent patterns hidden in my cognitive- and systems-neuroscientific data. It's a great, solidly built, and highly flexible package!" - Eelke Spaak