release of the Python library for scientific computing , focused on working with multidimensional arrays and matrices, as well as providing a large collection of functions with the implementation of various algorithms related to the use of matrices. NumPy is one of the most popular libraries used for scientific computing. The project code is written in Python using C optimizations and under the BSD license.
NumPy 1.19 dropped support for Python 3.5 and removed code to support Python 2 (numpy.compat layer left for now). Python 3.6, 3.7 and 3.8 are declared as supported versions. Module development continued to work with random samples. Improved support for NumPy wheel packages on the Aarch64 architecture and when using the Python implementation . numpy.frompyfunc, np.str_, numpy.copy, numpy.linalg.multi_dot, numpy.count_nonzero and numpy.array_equal functionality. Improved detection of CPU features such as AVX support. Added 5-7 times faster implementation based on AVX512, used for input data with type np.float64.
Source: opennet.ru