MindSpore
Design
MindSpore Design Overview
Programming Paradigm
Functional Differential Programming
MindSpore IR (MindIR)
Full-scenarios Unification
Combination of Dynamic and Static Graphs
Third-Party Hardware Interconnection
Distributed Training Design
Graph-Kernel Fusion Acceleration Engine
High Performance Data Processing Engine
Glossary
Models
Official Models
API
mindspore
mindspore.nn
mindspore.ops
mindspore.ops.primitive
mindspore.amp
mindspore.train
mindspore.communication
mindspore.common.initializer
mindspore.dataset
mindspore.dataset.transforms
mindspore.mindrecord
mindspore.nn.probability
mindspore.rewrite
mindspore.boost
mindspore.numpy
mindspore.scipy
API Mapping
PyTorch and MindSpore API Mapping Table
TensorFlow and MindSpore API Mapping Table
Migration Guide
Overview of Migration Guide
Environment Preparation and Information Acquisition
Model Analysis and Preparation
Constructing MindSpore Network
Debugging and Tuning
Network Migration Debugging Example
FAQs
Syntax Support
Static Graph Syntax Support
Tensor Index Support
FAQ
Installation
Data Processing
Implement Problem
Network Compilation
Operators Compile
Migration from a Third-party Framework
Performance Tuning
Precision Tuning
Distributed Parallel
Inference
Feature Advice
RELEASE NOTES
Release Notes
MindSpore
»
MindSpore Documentation
View page source
MindSpore Documentation
Design
The design concept of MindSpore's main functions to help framework developers better understand the overall architecture.
Model Libraries
Contains model examples and performance data for different domains.
API
MindSpore API description list.
API Mapping
API mapping between PyTorch and MindSpore, TensorFlow and MindSpore provided by the community.
Migration Guide
The complete steps and considerations for migrating neural networks from other machine learning frameworks to MindSpore.
Syntax Support
Syntax support for static graphs, Tensor indexes, etc.
FAQ
Frequently asked questions and answers, including installation, data processing, compilation and execution, debugging and tuning, distributed parallelism, inference, etc.
RELEASE NOTES
Contains information on major features and augments, API changes for the release versions.