Python API¶
Overview¶
This API section details functions, modules, and objects included in Apache MXNet, describing what they are and what they do. The APIs are grouped into the following categories:
Imperative API¶
MXNet NP module, an array library that provides NumPy-compatible API.
MXNet NPX module, neural network extension to the NumPy-compatible API.
Imperative APIs to load data, construct and train neural networks.
Advanced modules¶
API for querying MXNet enabled features.
MXNet array device for specifying in-memory storage device.
MXNet memory and performance profiler.
Managing symbolic graph execution.
Server node for the key value store.
Engine properties management.
Tools for compiling and running CUDA code from the python frontend.
Tools for using and testing MXNet.
General utility functions
Legacy¶
Imperative APIs to manipulate multi-dimensional arrays.
Symbolic APIs for multi-dimensional arrays and neural network layers
Functions to track various statuses during an epoch.
Image iterators and image augmentation functions.
Data iterators for common data formats and utility functions.
Read and write for the RecordIO data format.
Functions for Symbol visualization.