Standard Classifier

Recognize anything using large data collection

The process for creating a SDK package is a 5 step process. Step 3 is optional.

After having completed all 5 steps, you can download your first SDK package and integrate it together with the SDK code provided by Passio into your application. However, it is to be expected, that you might want to do a couple of iterations on some or each of the 5 steps before having created resources that satisfy your requirements.

Possible iteration cycles:

a) After having completed Step 1 through 5, you are testing the results of your SDK in the preview functionality of the Mobile app. You realize that certain labels are not yet detected with sufficient stability and decide to do some additional data collection for those labels, and train a new model.

b) After having completed Step 1 through 4, you realize that the test accuracy of your model is not yet good enough, and you want to merge some labels that are visually too similar to each other. After that, you will train a new model.

If at any point you need help from the Passio team, please reach out to us at [e-mail]