Use tf.keras.models.load_model instead of tf.saved_model.load

hunglc007 / tensorflow-yolov4-tflite

YOLOv4, YOLOv4-tiny, YOLOv3, YOLOv3-tiny Implemented in Tensorflow 2.0, Android. Convert YOLO v4 .weights tensorflow, tensorrt and tflite

MIT License

2.23k stars 1.24k forks source link

Closed toddwong closed 3 years ago

toddwong commented 3 years ago

Fix #282

cschar commented 3 years ago

seems a bit slower, but works correctly for all frames. Using yolov4-tiny-416, gives ~50ms per frame using mac m1