ERR:DeadlineExceeded

aivillain · December 14, 2021, 7:37am

On training mnist dataset image classification, error occurred.

ERR: DeadlineExceeded - Job was active longer than specified deadline Try decreasing the number of windows or reducing the number of training cycles. If the error persists then you can contact support at hello@edgeimpulse.com to increase this time limit.

I guess my project need to have more compute time.
how can i fix it?

louis · December 15, 2021, 8:39am

Hello @aivillain,

I can see that your dataset contains 42000 images for the training and that you are trying to classify 10 different classes with the MobileNetv2 0.05.

To stay within the free 20 minutes of compute per job, I would reduce first your dataset. Try first with something between 100 and 1000 images and increase it slowly until you reach the limits. Additionnally, you can try removing some classes and/or you can choose another Transfer Learning model (MobileNetv1 for example that are a lighter).

If your project is for another purpose than just testing with the MNIST dataset and you need it for a real use case, feel free to send me an email with the details of your project at louis@edgeimpulse.com and I can increase your time compute.

Regards,

Louis

cordic45 · July 18, 2022, 9:05am

@louis
Here, I have the same problem.
The purpose of my project is to detect 7 segments on an esp32-cam.
Although I try to reduce the dataset, the model does not perform well.

So
I would like to upgrade my account. But I cant’t find the page
or
Can I train FOMO on local computer?

shawn_edgeimpulse · July 19, 2022, 3:49pm

Hi @cordic45,

If it is for a sustainable cause, I can up your limit if you explain the details your project here.
If it is for a commercial project, I invite you to contact our sales team here: Pricing
If it is for testing purposes, you can use the tips outlined on this page: Lower compute time - Edge Impulse Documentation
Also note, if you are willing to share your project publicly, you can put a nice description on the your project’s description (https://www.edgeimpulse.com/blog/showcase-your-public-projects-with-markdown-readmes) with as much context as you can (goal of this project, etc.) and I can review it to increase your time limit.

cordic45 · July 20, 2022, 1:24am

hello, @shawn_edgeimpulse

I am just an intern, even this project is only meant for internal use. I am still unsure if I can opensource the company’s project.

What I’m going to do is use a MCU to recognize numbers on a meter. And sent the result to the server.

Currently, Raspberry Pi are used, but their cost is too high since thousands of small and large machines are in the Fab.