Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Create TensorFlow JS model to do inference #5

Closed
mattmotoki opened this issue Oct 23, 2018 · 3 comments · Fixed by #43
Closed

Create TensorFlow JS model to do inference #5

mattmotoki opened this issue Oct 23, 2018 · 3 comments · Fixed by #43
Assignees

Comments

@mattmotoki
Copy link
Contributor

mattmotoki commented Oct 23, 2018

Setup a TensorFlowJS model to do inference in the browser

  • load pretrained weights
  • make predictions on new images
@mattmotoki mattmotoki changed the title Create TensorFlow Lite model to do inference Create TensorFlow JS model to do inference Oct 24, 2018
@rohit-basavaraju rohit-basavaraju self-assigned this Oct 29, 2018
@rohit-basavaraju
Copy link
Contributor

Looks like there are streamlined steps to export keras model written in python to tensorflow.js! https://js.tensorflow.org/tutorials/import-keras.html

@rohit-basavaraju rohit-basavaraju removed their assignment Oct 29, 2018
@mattmotoki
Copy link
Contributor Author

Adding to that, this would be how you could take a pretrained MobileNet and save it in a format that tensorflow.js can use.

import tensorflowjs as tfjs
from keras.applications.mobilenet import MobileNet

model = MobileNet(weights='imagenet')
tfjs.converters.save_keras_model(model, 'test_model')

@mrbarbasa
Copy link
Contributor

Related to #29.

@mrbarbasa mrbarbasa self-assigned this Oct 29, 2018
@mrbarbasa mrbarbasa added the in progress Work is in progress label Oct 29, 2018
@mrbarbasa mrbarbasa removed the in progress Work is in progress label Nov 1, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants