Why jetson nano fp16 is slower than fp32

530869411 · May 23, 2021, 2:24pm

when i use fp16 to inference is slower than fp32

kayccc · May 24, 2021, 1:48am

Topic		Replies	Views
Why inference in jetson nano with fp16 is slower than fp32 Jetson Nano tensorrt , jetson-inference	9	1926	September 5, 2021
Jetson Nano 16bits vs 32 bits inference performance Jetson Nano tensorrt , jetson-inference , python	2	646	April 18, 2023
Faster inference in tensorrt model TensorRT tensorrt , jetson-inference	1	395	April 3, 2023
Does nano deep learning support fp16 Jetson Nano jetson-inference	2	543	October 15, 2021
Why jetson nano not support int8 Jetson Nano	4	4987	October 14, 2021
About the time taken for FP16 inference and int8 inference in TensorRT TensorRT tensorrt	1	342	November 15, 2023
FP16 shows slower inference on Turing vs FP32 Frameworks cuda	0	623	January 6, 2021
FP16 mode is not running faster than FP32 mode TensorRT	0	930	February 11, 2019
Does training deep learning mode on fp16 uses less memory? Jetson Nano ai-training	4	573	October 15, 2021
Model inference speed suddenly slows down Jetson Nano	2	529	October 14, 2021