Statements (19)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:Neural_Engine
|
| gptkbp:accuracy |
floating point
|
| gptkbp:announced |
2017
|
| gptkbp:architecture |
ASIC
|
| gptkbp:developedBy |
gptkb:Google
|
| gptkbp:formFactor |
PCIe card
|
| gptkbp:memoryPerDevice |
8 GB HBM
|
| gptkbp:performanceContext |
45 teraflops
|
| gptkbp:predecessor |
gptkb:TPU
|
| gptkbp:successor |
TPU3
|
| gptkbp:supports |
gptkb:TensorFlow
|
| gptkbp:usedFor |
gptkb:machine_learning
deep learning neural network training neural network inference |
| gptkbp:usedIn |
gptkb:Google_Cloud_Platform
|
| gptkbp:bfsParent |
gptkb:Time_Processor_Unit_(TPU)
|
| gptkbp:bfsLayer |
8
|
| https://www.w3.org/2000/01/rdf-schema#label |
TPU2
|