What do you think about the reportet below sentences?
In this case, there is no alternative to Python - not [only] because it better at ML, but also because of the array of solutions and tools at your service. The only reason to use Golang for ML is a (theoretical) speedup - but all the math in Python ML stack is already written in C! (not to mention Spark, etc). It does not mean it is impossible to beat python ML with Golang (Go, for example, leverages multithreading by default, AFAIK) - it means that you need to be an expert on Data Structures, Data Science and Go language [read: seasoned ML expert with a clear understanding of the reasons and consequences] to do that