Skip to content
This repository has been archived by the owner on Mar 19, 2023. It is now read-only.

QUESTION: How secure is it to train with sensitive images? Is it all done locally? #49

Open
pain0486 opened this issue Dec 8, 2021 · 2 comments

Comments

@pain0486
Copy link

pain0486 commented Dec 8, 2021

I didn't see any uploading in the network tabs and have not found anything yet in the code where the images were being uploaded anywhere. Is the training all done locally?

I was wanting to train the model to detect blurry images of documents, but the training material has personally identifying information and I want to me 100% sure its safe

@insaurabh
Copy link

Hi @pain0486

I was searcing the same in the code level. Do you found anything related to this.

Thanks.

@HalfdanJ
Copy link
Contributor

No images are at any point uploaded to any server when you train or do inference. All happens clientside in tensorflow.js.

Only situations that teachable machine uploads images is if you choose to save your project on Google drive, or if you use the export converter for quantized models (the converter needs a few example images to do that operation) but those photos are not stored anywhere.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants