Statements (17)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
Neural Engine
|
| gptkbp:announced |
gptkb:Google_I/O_2016
|
| gptkbp:architecture |
ASIC
|
| https://www.w3.org/2000/01/rdf-schema#label |
TPU1
|
| gptkbp:manufacturer |
gptkb:Google
|
| gptkbp:memoryType |
8 GB DDR3
|
| gptkbp:notableFor |
first generation TPU
|
| gptkbp:performanceContext |
92 TOPS
|
| gptkbp:powerSource |
40W
|
| gptkbp:purpose |
accelerate machine learning workloads
|
| gptkbp:releaseYear |
2015
|
| gptkbp:successor |
gptkb:TPU2
|
| gptkbp:supports |
8-bit integer arithmetic
|
| gptkbp:usedFor |
gptkb:Google_data_centers
gptkb:TensorFlow |
| gptkbp:bfsParent |
gptkb:Time_Processor_Unit_(TPU)
|
| gptkbp:bfsLayer |
8
|