MindSpore is the AI computing framework
MindSpore Studio provides the following features:
*User-friendly NPU-based programming GUI
Operator developers can customize CCE development on MindSpore Studio based on
the CCE programming depth to implement in-depth integration. The keywords of the
extended CCE language are highlighted. You can compile heterogeneous hybrid codes in
one-click mode.
*NPU-based graphical debugging
For the development of the operator acceleration library on the NPU, MindSpore Studio
provides a graphical GUI for users to implement real-time tracking of the running status
of the acceleration operators on the AI core and AI CPU.
*Automatic offline model management
Trained third-party models, such as Caffe models, can be imported to MindSpore Studio
and converted into models supported by the system. Model interfaces are generated
automatically in one-click mode, enabling interface-based model programming. For
details, see 4 "Offline Model Conversion" in the Ascend 310 MindSpore Studio Quick
Start.
* "Zero" programming for service process orchestration
For service process developers, MindSpore Studio provides the drag-and-drop
programming mode based on service nodes. You can implement service orchestration by
simply dragging and connecting service nodes. The one-stop service after orchestration,
ranging from compilation and running to result display, makes process development
smarter. "Zero" programming is involved. In this way, you can get started quickly
without extra learning costs. For details, see 3 "Building the First Machine Learning
App" in the Ascend 310 MindSpore Studio Quick Start .
*Graphical TE programming
MindSpore Studio provides the industry's first integrated development environment
based on the TVM-based Tensor Engine (TE) for programming development. Operators
can be transplanted quickly across platforms, enabling instant NPU adaption.
*Log Analysis
MindSpore Studio provides a system-wide log collection and analysis solution for the
NPU platform, improving the efficiency of locating runtime algorithm problems. A
unified log format is adopted. Visualized analysis of cross-platform logs and runtime
diagnosis runs in Web mode, improving the usability of the log analysis system.
*Performance analysis
MindSpore Studio provides graphical user interfaces (GUIs) and command-line
interfaces (CLIs) to implement efficient, easy-to-use, and flexible performance profiling
on the multi-node and multi-module heterogeneous system on the host and device.
Synchronous analysis of performance and power consumption of the NPU device is
implemented, which meets the requirements of algorithm optimization for system
performance analysis.
* Simulation
Function-level simulation execution libraries for the AI core under the Caffe framework
are provided. You can call AI core simulation by using the program.