Statements (58)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:software
|
| gptkbp:bfsLayer |
4
|
| gptkbp:bfsParent |
gptkb:NVIDIA_Triton_Inference_Server
|
| gptkbp:developed_by |
gptkb:Microsoft
|
| gptkbp:has |
community support
performance benchmarks active development community |
| https://www.w3.org/2000/01/rdf-schema#label |
ONNX Runtime
|
| gptkbp:integrates_with |
gptkb:Azure_Machine_Learning
|
| gptkbp:is |
cost-effective
flexible lightweight open-source widely adopted efficient modular user-friendly robust highly scalable extensible reliable secure platform agnostic performance-oriented future-proof |
| gptkbp:is_available_on |
gptkb:archive
|
| gptkbp:is_compatible_with |
multiple hardware platforms
|
| gptkbp:is_documented_in |
official documentation
|
| gptkbp:is_integrated_with |
gptkb:Graphics_Processing_Unit
gptkb:scikit-learn gptkb:Py_Torch |
| gptkbp:is_optimized_for |
model performance
AMDGP Us NVIDIAGP Us Intel CP Us |
| gptkbp:is_part_of |
ONNX ecosystem
|
| gptkbp:is_used_by |
gptkb:physicist
gptkb:software data scientists |
| gptkbp:is_used_for |
machine learning model inference
|
| gptkbp:is_used_in |
cloud environments
edge devices |
| gptkbp:language |
gptkb:Java
gptkb:C++ gptkb:Library gptkb:C# |
| gptkbp:offers |
cross-platform support
|
| gptkbp:provides |
model conversion tools
custom operators API for model management high-performance inference engine |
| gptkbp:released_in |
gptkb:2019
|
| gptkbp:supports |
quantization
multi-threading multiple frameworks model pruning Open Neural Network Exchange (ONNX) format |
| gptkbp:written_in |
gptkb:C++
|