I have trained a model using Roboflow work space. The model works fine with web cam web interface.
When I try running the inference on OAK-1 from below code. I can see the stream but no predictions found. I would appreciate any inputs regarding this.
We pushed a fix for our OAK deployment yesterday. Can you confirm that you’re using version 0.0.11 (or later)?
You can update using pip:
pip install roboflowoak==0.0.11
Hi Leo
I confirm you that I installed the version roboflowoak==0.0.11, but it generated the following error
Traceback (most recent call last):
File “C:\Users\ingca\AppData\Local\Programs\Python\Python311\Lib\site-packages\roboflowoak_init_.py”, line 72, in init
self.dai_pipe = DepthAIPipeline(
^^^^^^^^^^^^^^^^
File “C:\Users\ingca\AppData\Local\Programs\Python\Python311\Lib\site-packages\roboflowoak\pipe.py”, line 87, in init
self.cam_rgb.out.link(self.detection_nn.input)
^^^^^^^^^^^^^^^^
AttributeError: ‘depthai.node.ColorCamera’ object has no attribute ‘out’
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File “C:\Users\ingca\AppData\Local\Programs\Python\Python311\Lib\site-packages\roboflowoak_init_.py”, line 104, in init
self.dai_pipe = DepthAIPipeline(
^^^^^^^^^^^^^^^^
File “C:\Users\ingca\AppData\Local\Programs\Python\Python311\Lib\site-packages\roboflowoak\pipe.py”, line 87, in init
self.cam_rgb.out.link(self.detection_nn.input)
^^^^^^^^^^^^^^^^
AttributeError: ‘depthai.node.ColorCamera’ object has no attribute ‘out’
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File “C:\Users\ingca\OneDrive\Documentos\Proyecto\Proyecto.py”, line 8, in
rf = RoboflowOak(model=“trabajo-de-grado-final”, confidence=0.05,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “C:\Users\ingca\AppData\Local\Programs\Python\Python311\Lib\site-packages\roboflowoak_init_.py”, line 123, in init
raise Exception(
Exception: Failure while retrying load weights - does this model, version, api key exist? can you curl api.roboflow.com, and can your device download files from google cloud storage? have you hit your device limit?
from roboflowoak import RoboflowOak
import cv2
import time
import numpy as np
if name == ‘main’:
# instantiating an object (rf) with the RoboflowOak module
rf = RoboflowOak(model=“trabajo-de-grado-final”, confidence=0.05,
overlap=0.5, version=“2”,
api_key=“aDBlDcA4YIxszVDvS6WH”, rgb=True, depth=False,
device=None, device_name=None, blocking=True)
while True:
t0 = time.time()
result, frame, raw_frame, depth = rf.detect(visualize=True)
predictions = result[“predictions”]
#{
# predictions:
# [ {
# x: (middle),
# y:(middle),
# width: ,
# height: ,
# depth: ###->,
# confidence: ,
# class: ,
# mask: { }
# }
# ]
#}
#frame - frame after preprocs, with predictions
#raw_frame - original frame from your OAK
#depth - depth map for raw_frame, center-rectified
# to the center camera
# To access specific values within “predictions” use:
# p.json()[a] for p in predictions
# set “a” to the index you are attempting to access
# Example: accessing the “y”-value:
# p.json()[‘y’] for p in predictions
t = time.time()-t0
print("INFERENCE TIME IN MS ", 1/t)
print("PREDICTIONS ", [p.json() for p in predictions])
# setting parameters for depth calculation
#max_depth = np.amax(depth)
#cv2.imshow("depth", depth/max_depth)
# displaying the video feed as successive frames
cv2.imshow("frame", frame)
# how to close the OAK inference window/stop inference:
# CTRL+q or CTRL+c
if cv2.waitKey(1) == ord('q'):
break
I have been following this thread as I’ve also been experiencing the same failure as @Sergio_Castellanos (and others) to get inferencing running on an Oak device.
I can add that I am getting the identical exceptions/errors as reported by @Sergio using roboflowoak 0.0.11
@Sergio_Castellanos: I’m investigating your issue, but in the meantime, I’ve revoked the use of your API key, which you published in your post. You can generate a new one in your settings. In the future, please be careful to post information that could give others unauthorized access to your accounts.
Hi Leo an Larry , thanks for your support and solution, I confirm that when I install the version roboflowoak==0.0.12 and run the script, the roboflow integration with my oak -1 camera works perfectly.
great job
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.