|
|
|
|
|
|
by twtw
3042 days ago
|
|
|
IIRC, TPUv2 uses 16 bit floating point in some format with higher dynamic range and lower precision than standard fp16. Can someone confirm? If that is right, is the "Tensorflow-optimized" Resnet-50 using 16bit floats when running on TPUv2? |
|
|