In Rappi as in many other high-potential startups, it is clear that one of the keys to success has been and continues to be the implementation of analytics and data science, using machine learning models that provide valuable insights to the business. Its use in startups and traditional companies that have focused on digital transformation has been increasing exponentially and today, being a part of the broader AI field, machine learning should be as common as software applications in general, and that is precisely where MLOps treat ML algorithms as reusable software appliances, offering rapid and repeatable deployment of models, followed by continuous and monitored integration ensuring that each model performs optimally as its environment evolves over time. In other words, and to wrap up, MLOps are the set of practices that an enterprise must have in place in order to run AI and ML successfully. If you have data science and IA you must have almost by obligation a dedicated MLOps team, the models by themselves are helpless and biased only to the data they were trained with. Nowadays, a start-up must face big challenges in terms of managing their data as it is constantly growing and changing.

After extensively researching the best software for face recognition, I came to the conclusion that almost all the articles currently published are just copied and pasted from advertisements. Even worse, most of these articles recommend outdated libraries and services that are not supported anymore. Some of their suggested solutions can't even run on modern operating systems! I promise, this is not one of those articles. I've done my best to make a comprehensive list of all the modern face recognition solutions on the market.

Statsmodels is an open-source statistics-driven module that offers various classes and functions to the many statistical models available for statistical analysis and exploration of data. The module covers a vast number of models ranging from Linear Regression, Discrete Models, Time Series Analysis, Survival Analysis, and many other miscellaneous models.

CausalNex is a Python library that allows data scientists and domain experts to co-develop models that go beyond correlation and consider causal relationships. 'CasualNex' provides a practical'what if' library which is deployed to test scenarios using Bayesian Networks (BNs). 'CasualNex' prepares practitioners to understand structural relationships from data and helps in the verification for accuracy of the relationships between different data sets. Apart from practitioners understanding the structural relationship from data, it also enables domain experts to fit conditional probability distributions and study the effect of potential interventions. 'CasualNex' helps to simplify the following steps: CausalNex is a Python package.

stream-learn is a Python package compatible with scikit-learn and developed for the drifting and imbalanced data stream analysis. I ts main component is a stream generator, which allows to produce a synthet ic data stream that may incorporate each of the three main concept drift typ es (i.e. The package allows conducting experiments following estab lished evaluation methodologies (i.e. In addition, estimators adapted for data stream classification have been implem ented, including both simple classifiers and state-of-art chunk-based and online classifier ensembles. To improve computational efficiency, package utili ses its own implementations of prediction metrics for imbalanced binary cla ssification tasks. Keywords: Data stream, Concept drift, Imbalanced data, Dynamic class imbalance 1. Motivation and significance Pattern recognition research increasingly goes beyond the usual pattern of building classification models on stationary data sets an d focuses on data stream processing where class distributions, and hence als o decision boundaries, may change over time [1].