python == 3.10.14
cuda == 12.0
torch == 2.1.1+cu121
torchaudio == 2.1.1+cu121
torchvision == 0.16.1+cu121
ultralytics == 8.2.101
tf_keras == 2.17.0
tflite-support == 0.4.4
tifffile == 2024.7.24
The assets folder contains tflite files converted using the YOLOV8 project and corresponding tag txt files
The result folder contains the inference results of the same image using the same model on different platforms. The result_comuter is the result of using YOLOV8 to load best_float32.tflite inference cls_person_2. jpg on the computer Result_phone.jpg is the result of inferring cls_person_2. jpg on a computer or Android phone
The release folder contains the apk installation package built using the best_float32.tflite model
The weights folder contains YOLOV8 trained PT files and converted files in other formats