Dorotea Jalušić
Dorotea Jalušić
No, I use DHCP. When I use Auto IP, the camera is not found in PyPylon (the function returns an empty tuple almost instantly). I copied this information from the...
I tried increasing the interpacket delay multiple times to higher values. I also changed image format from YCbCr to Bayer RG8 to decrease the payload. However, the period of frames...
Sorry it's taking so long, I managed to capture the traffic in the next occurence, but it seems it doesn't have all the information as the network infrastructure is more...
Can I contact you via email? I can share the file with captured packets, but it may not contain all the necessary information (UDP protocol).
The serial numbers are: 2823671257, 2823671269, 2823670490, 2823671265.
@ricardobnjunior, have you succeeded in converting the model to TFLite?
According to this [link](https://stackoverflow.com/questions/54420363/multi-class-object-detection-with-tensorflow-weird-behavior-on-evaluation), the filenames have to be shuffled inside the generate_tfrecords.py in order to properly train the model to detect multiple classes.
TensorFlow now has a new runtime framework for on-device AI, [LiteRT](https://ai.google.dev/edge/litert), so TFLite (and TFLite Model Maker) is no longer maintained. If you try to use these tutorial notebooks in...