Slide 5
Slide 5 text
Performance
TF Lite is fast with no noticeable accuracy loss
Portability
Android, iOS, and more specialized IoT devices.
Low latency
Optimized float- and fixed-point CPU kernels, op-fusing, and more.
Acceleration
Integration with GPU and internal/external accelerators.
Small model size
Controlled dependencies, quantization and op registration.
Tooling
Conversion, compression, benchmarking, power-consumption
휴대용 모바일 기기(Portable)에서
적은 지연(Low latency)과
작은 모델 사이즈(Small Model Size)로
OnDevice 딥러닝 추론을 가능하게 해주는
머신러닝 프레임워크