دانلود کتاب تیم های یادگیری ماشین موثر

Whether you’re a ML practitioner in academia, enterprise organizations, start-ups, scale-ups, or consulting, the principles and practices in this book can help you and your team identify opportunities for improvement and be more effective at what you do. Data scientists and ML engineers The job scope of a data scientist has evolved over the past few years. Instead of purely focusing on modeling techniques and data analysis, we’re seeing expectations (implicit or explicit) that one needs the capabilities of a full-stack data scientist: data wrangling, ML engineering, MLOps, business case formulation, among others. This book will elaborate on the essential capabilities (beyond ML) which are required in designing and delivering ML solutions in the real world. We have presented the principles, practices and hands-on exercises in this book to various groups of ML practitioners (data scientists, ML engineers, PhD students, software engineers) over several years, and we’ve consistently received positive feedback. The ML practitioners that we have worked with in the industry said that they benefited from improvement in feedback cycles, flow and reliability that comes from practices such as automated testing and refactoring. Our takeaway from that is that there is a desire from the ML community to learn these skills and practices, and this is our attempt to scale the sharing of this knowledge. Software engineers and infrastructure engineers When we run workshops on this topic, we have noticed another persona or class of people who are working in the ML space – and that is software engineers, infrastructure engineers, or platform engineers. While capabilities from the software world (e.g. infrastructure as code, deployment automation, automated testing, etc.) are necessary in designing and delivering ML solutions in the real-world, they are not sufficient. To create reliable ML solutions, we need to be able to properly see the entire problem space. To do so, we will supplement the software lens of looking at the world with other principles and practices, such as ML model tests, dual-track delivery, shifting ethics left, among others, to handle challenges that are unique to ML.