Increase Deep Learning Framework Performance on CPUs and GPUs
Building Blocks to Optimize AI Applications
The Intel® oneAPI Deep Neural Network Library (oneDNN) helps developers improve productivity and enhance the performance of their deep learning frameworks. Use the same API to develop for CPUs, GPUs, or both. Then implement the rest of the application using Data Parallel C++. This library is included in both the Intel® oneAPI Base Toolkit and Intel® oneAPI DL Framework Developer Kit.
The library is built around three concepts:
Primitive: Any low-level operation from which more complex operations are constructed, such as convolution, data format reorder, and memory
Engine: A hardware processing unit, such as a CPU or GPU
Stream: A queue of primitive operations on an engine
Supports key data type formats, including 16- and 32-bit floating points, bfloat16, and 8-bit integers
Implements rich operators, including convolution, matrix multiplication, pooling, batch normalization, activation functions, recurrent neural network (RNN) cells, and long short-term memory (LSTM) cells
Accelerates inference performance with automatic detection of Intel® Deep Learning Boost technology
Get what you need to build, test, and optimize your oneAPI projects for free. With an Intel® DevCloud account, you get 120 days of access to the latest Intel® hardware—CPUs, GPUs, FPGAs—and Intel oneAPI tools and frameworks. No software downloads. No configuration steps. No installations.