Functional Programming Paradigm of Python for Scientific Computation Pipeline Integration

As an interpreted programming language, Python is of the characteristics of concise syntax, flexibility on multiple programming paradigms, cross-platforms, etc. [1-3]. Its software ecosystem gets enriched, which benefits from the contributions of soaring researchers and developers studied in various fields. As with the booming development of AI techniques generally, Python has become a de facto development standard for scientific computation and AI algorithms due to not only some excel high performance libraries such as numpy, scipy, and tensorly [4-6], but also high compatibility for integrating other programming languages. Nonetheless, as the expense of exceeding flexibility and the richness of software ecosystem, it gets also challenging in integration for complicated project via Python, particularly in scientific computation used in interdisciplinary application [7-10]. Such as, most of data manipulations are generally in demands of importing third party libraries designed on basis of different specifications, which can frequently result in incompatibility problems raised by data types or such like.