[XLA:GPU] Implement BatchNormThunk as a call into cudnn.
Using cudnn for these calls is disabled by default, because it's not a performance win on our benchmarks. PiperOrigin-RevId: 179882911
Loading
Please sign in to comment
Using cudnn for these calls is disabled by default, because it's not a performance win on our benchmarks. PiperOrigin-RevId: 179882911