Closed bhavikapanara closed 3 years ago
@bhavikapanara you've copied virtually all the code from the ambianic-edge repo. This is not at all the goal here.
Please remove all files and refactor fall detection classes as needed. If there is any code that is not directly involved in the fall detection algorithm, then it does not belong in this repo.
You should also have a very clear plan and design how ambianic-edge will import and use the fall-detection python package.
Please prepare a design document so we can align on high level goals. I usually use diagrams.net for design diagrams but you can use any similar tool that you like.
@ivelin Yes...I have tried to consider the only code that is directly involved in fall detection. However, again I will recheck it and refactor fall detection classes if require.
And, Will prepare the design document and get back to you.
Check out this pull request on
See visual diffs & provide feedback on Jupyter Notebooks.
Powered by ReviewNB
@bhavikapanara I don't see a README.md for the repo. Why was it deleted?
@bhavikapanara I don't see a README.md for the repo. Why was it deleted?
Ohh...by mistake it deleted. you can see it in the next update.
@bhavikapanara commented on a few more areas that need work.
tried to cover all comments in the latest update...Please review it and let me know if I missed anything.
See additional comments.
Please comment how you think about using this standalone library in ambianic-edge. What is the API that will be used by edge for this model and potentially other standalone AI models?
we can use the Fall_prediction
method (exist in fall_prediction.py)
with pipeline image 2 or 3. Similar to the demo example demo-fall-detection.py
. I think this is one way to call this fall-detect stand-alone package in ambianic-edge.
@ivelin Please guide me if any other good method that we can implement.
OK, can you please show me a few lines of pseudo code. Which class in the ambianic edge will interact with the fall detector package?
On Tue, Jun 1, 2021 at 10:38 PM bhavika panara @.***> wrote:
See additional comments.
Please comment how you think about using this standalone library in ambianic-edge. What is the API that will be used by edge for this model and potentially other standalone AI models?
we can use the Fall_prediction method (exist in fall_prediction.py) with pipeline image 2 or 3. Similar to the demo example demo-fall-detection.py. I think this is one way to call this fall-detect stand-alone package in ambianic-edge.
Please guide me if any other good method that we can implement.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ambianic/fall-detection/pull/14#issuecomment-852691449, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFKPVWOYU2BDFD6ISHDTQWRTFANCNFSM43KLVPMQ .
after brainstorming with the goal I found one possible way to integrate an ambianic-edge with the fall-detect module.
We can directly attach an ambianic-edge pipeline with a fall-detect module and call the fall_detection process on the receive_next_sample()
method of PipeElement
def receive_next_sample(self, **sample):
Here is an example to call fall-detect module: Link
after brainstorming with the goal I found one possible way to integrate an ambianic-edge with the fall-detect module.
We can directly attach an ambianic-edge pipeline with a fall-detect module and call the fall_detection process on the
receive_next_sample()
method ofPipeElement
def receive_next_sample(self, **sample):
Here is an example to call fall-detect module: Link
Yes, that seems like a sensible approach.
@bhavikapanara OK, let's give it a shot. Please be on standby for potential issues we have not caught in the review process.
yes...sure @ivelin Thanks
:tada: This PR is included in version 1.0.0 :tada:
The release is available on GitHub release
Your semantic-release bot :package::rocket:
This PR is initialization of stand-alone fall detect python package
Run python file for fall-detection:
python3 demo-fall-detection.py
To test fall-detection using the command line for 2 images:
python3 demo-fall-detection-cmd.py --image_1 Images/fall_img_1.png --image_2 Images/fall_img_2.png
To test fall-detection using the command line for 3 images:
python3 demo-fall-detection-cmd.py --image_1 Images/fall_img_1.png --image_2 Images/fall_img_2.png --image_3 Images/fall_img_3.png
Use
Demo.ipynb
jupyter-notebook for the experiment.