API

MindSpore provides rich interfaces for model building, training, and inference. The functions of each module interface are described below.

Module Name	Descriptions
mindspore	Framework foundation interface.
mindspore.nn	A neural network layer for building predefined building blocks or computational units in a neural network.
mindspore.ops	Function interface.
mindspore.ops.primitive	Primitive operator.
mindspore.mint	Functional, nn, and optimizer interfaces consistent with mainstream industry usage.
mindspore.amp	Mixed-precision interface.
mindspore.train	Traning interface.
mindspore.communication	Collection communication interface.
mindspore.communication.comm_func	Collection communication functional interface.
mindspore.common.initializer	Parameter initialization.
mindspore.hal	Interface of device management, stream management, event management and memory management.
mindspore.dataset	Interfaces for loading and processing various datasets.
mindspore.dataset.transforms	Generalized data transformations.
mindspore.mindrecord	Operations interface related to efficient data format MindRecord developed by MindSpore.
mindspore.nn.probability	Parameterizable probability distributions and sampling functions.
mindspore.rewrite	Custom rule-based model source code modification interface.
mindspore.multiprocessing	Multi-processing interface.
mindspore.boost	Automatic acceleration network interfaces.
mindspore.numpy	NumPy Class interface.
mindspore.scipy	SciPy Class interface.
mindspore.utils	Tools interface.
mindspore.experimental	Experimental interface.
Environment Variables	Notes related to environment variables.