Statements (56)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
software
|
| gptkbp:availableIn |
Linux
Windows Jetson platforms |
| gptkbp:canEstablish |
TensorFlow models
PyTorch_models Keras_models |
| gptkbp:compatibleWith |
gptkb:ONNX
|
| gptkbp:developedBy |
NVIDIA
|
| gptkbp:features |
dynamic tensor memory
layer fusion precision calibration kernel auto-tuning |
| gptkbp:has |
community support
low latency high throughput extensive documentation |
| gptkbp:hasVersion |
8.4.1
|
| https://www.w3.org/2000/01/rdf-schema#label |
TensorRT
|
| gptkbp:integratesWith |
CUDA
cuDNN |
| gptkbp:isCompatibleWith |
TensorFlow Serving
NVIDIA Triton Inference Server |
| gptkbp:isCounteredBy |
neural network models
|
| gptkbp:isOptimizedFor |
real-time applications
edge computing cloud inference |
| gptkbp:isPartOf |
NVIDIA AI platform
NVIDIA Deep Learning SDK |
| gptkbp:isSupportedBy |
NVIDIA forums
NVIDIA GitHub repositories NVIDIA_developer_tools |
| gptkbp:isUsedIn |
financial services
gaming industry robotics autonomous vehicles healthcare_applications |
| gptkbp:performance |
NVIDIA GPUs
|
| gptkbp:provides |
API for developers
profiling tools model optimization tools high-performance inference |
| gptkbp:publishedIn |
gptkb:C++
|
| gptkbp:releasedIn |
2016
|
| gptkbp:supports |
multi-GPU configurations
custom layers FP16 precision INT8 precision deep learning inference |
| gptkbp:usedBy |
data scientists
software engineers AI researchers |
| gptkbp:usedFor |
computer vision
natural language processing speech recognition |
| gptkbp:usedIn |
machine learning applications
|