I have update HailoRT and Tapps to the lastest version, but it can not run any example.
I installed hailo example, and I run object detection:
(hailo_platform_venv) ain@ain:~/object_detection $ ./object_detection.py -n ./yolov8s.hef -i output_image0.jpg
Traceback (most recent call last):
File "/home/ain/object_detection/./object_detection.py", line 18, in <module>
from utils import HailoAsyncInference, load_input_images, validate_images, divide_list_to_batches
ImportError: cannot import name 'HailoAsyncInference' from 'utils' (/home/ain/hailo_platform_venv/lib/python3.11/site-packages/utils/__init__.py)
When I use video stream
(hailo_platform_venv) ain@ain:~/streaming $ ./yolox_stream_inference.py
[HailoRT] [warning] HEF was compiled for Hailo8L device, while the device itself is Hailo8. This will result in lower performance.
[HailoRT] [warning] HEF was compiled for Hailo8L device, while the device itself is Hailo8. This will result in lower performance.
Traceback (most recent call last):
File "/home/ain/streaming/./yolox_stream_inference.py", line 67, in <module>
layer_from_shape: dict = {infer_results[key].shape:key for key in infer_results.keys()}
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/ain/streaming/./yolox_stream_inference.py", line 67, in <dictcomp>
layer_from_shape: dict = {infer_results[key].shape:key for key in infer_results.keys()}
^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'list' object has no attribute 'shape'
I see, that makes sense now. You can’t run this because PyHailoRT version 4.17 isn’t installed. We’ve already prepared a one-command installation, but we’re currently waiting for the Raspberry Pi release.
In the meantime, please check out my guide here: Upgrade HailoRT to 4.18 on RPi. If you have any further questions, feel free to ask!
Note : if you are going to use Python API only and not the hailo_rpi5_examples ( gstreamer ) you don’t need to install tappas !