-
### Description
Hi! When I try connect to my devboard via "mdt shell," I get the following error message:
I followed the instructions at https://coral.ai/docs/dev-board/mdt/#install-mdt, and it l…
-
What steps will reproduce the problem?
Have remote Debian based machine with active display (tested on https://coral.ai/products/dev-board/)
Set up remote interpreter via SSH in PyCharm (2021.1) /…
-
### Description
Hello,
I've come across an issue while trying to get started with the Coral USB accelerator. When I've tried running the example code it gave me an error saying that "/usr/lib/lib…
-
### Description
Hi,
I am trying to leverage TPU Edge HW on Raspberry Pi 4 Model B.
I trained an SSD Mobilenet 320x320 object detection model to run inference on RPI with TPU Edge.
I got better …
-
### Description
I have a EfficientDet Lite 3 model that I have retrained **with TRANSFER LEARNING** (code below), that I am attempting to compile using the Google Colab notebook provided by Google …
ghbn8 updated
2 years ago
-
### Description
Hi, I want to look at the code that runs when I runthe command `edgetpu_detect_server`. I just want to see how it streams usb camera to the Google browser. Thank you.
Click to expan…
-
**System information**
- Have I written code (based on the docs):
```
from tflite_runtime.interpreter import Interpreter
from tflite_runtime.interpreter import load_delegate
model_path='my_compil…
-
Hi, do you have any resources on how to train a new segmentation model to run on the Coral? (something like the Colab training tutorials?) I see the pre-trained models here https://coral.ai/models/sem…
-
### Description
Hi, I'm running into a memory leak when the edgetpu is runs in a thread.
from threading import Thread
import cv2
import numpy as np
from pycoral.utils import edg…
-
After my quick hacking to update to use pycoral the bounding boxes are incorrect - perhaps since the image is resized to (300, 300) before processing?