one API Deep Neural Network Library
GPTKB entity
Statements (72)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:Library
|
| gptkbp:developed_by |
gptkb:Intel
|
| gptkbp:enhances |
AI workloads
|
| gptkbp:facilitates |
model deployment
AI model development |
| https://www.w3.org/2000/01/rdf-schema#label |
one API Deep Neural Network Library
|
| gptkbp:includes |
matrix operations
performance benchmarks activation functions tensor operations convolution operations normalization layers DNNLAPI recurrent operations |
| gptkbp:integrates_with |
gptkb:Intel_one_API_Toolkits
|
| gptkbp:is_available_on |
gptkb:smartphone
gptkb:operating_system |
| gptkbp:is_compatible_with |
multiple hardware architectures
|
| gptkbp:is_designed_for |
performance optimization
|
| gptkbp:is_optimized_for |
gptkb:Intel_architectures
|
| gptkbp:is_used_for |
computer vision
natural language processing reinforcement learning inference training neural networks |
| gptkbp:offers |
gptkb:document
GPU acceleration community support CPU acceleration |
| gptkbp:part_of |
gptkb:one_API
|
| gptkbp:provides |
debugging tools
version control user-friendly interfaces pre-trained models data parallelism profiling tools API compatibility performance tuning options memory management features multi-threading capabilities high-performance primitives layer fusion capabilities layered AP Is |
| gptkbp:released_in |
gptkb:2020
|
| gptkbp:supports |
gptkb:Graphics_Processing_Unit
gptkb:CEO gptkb:Py_Torch deep learning model evaluation heterogeneous computing model training asynchronous execution cloud deployment custom kernels distributed training real-time inference Python API dynamic computation graphs C++ API model quantization CAPI F P16 precision B F16 precision multi-device execution sparse data formats |
| gptkbp:used_in |
gptkb:academic_research
gptkb:Research_Institute industry applications commercial products |
| gptkbp:written_in |
gptkb:C++
|
| gptkbp:bfsParent |
gptkb:Intel_One_API
|
| gptkbp:bfsLayer |
5
|