Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Import pretrained models from keras.applications

See original GitHub issue

TensorFlow.js version 0.6.1

Browser version Google Chrome | 65.0.3325.181

Describe the problem

Uncaught (in promise) Error: Unknown layer: SeparableConv2D
    at new t (tfjs@0.6.1:1)
    at Object.n.deserializeKerasObject (tfjs@0.6.1:1)
    at Object.n.deserialize (tfjs@0.6.1:1)
    at p (tfjs@0.6.1:1)
    at t.fromConfig (tfjs@0.6.1:1)
    at Object.n.deserializeKerasObject (tfjs@0.6.1:1)
    at Object.n.deserialize (tfjs@0.6.1:1)
    at tfjs@0.6.1:1
    at r (tfjs@0.6.1:1)
    at Object.next (tfjs@0.6.1:1)
t @ tfjs@0.6.1:1
n.deserializeKerasObject @ tfjs@0.6.1:1
n.deserialize @ tfjs@0.6.1:1
p @ tfjs@0.6.1:1
t.fromConfig @ tfjs@0.6.1:1
n.deserializeKerasObject @ tfjs@0.6.1:1
n.deserialize @ tfjs@0.6.1:1
(anonymous) @ tfjs@0.6.1:1
r @ tfjs@0.6.1:1
(anonymous) @ tfjs@0.6.1:1
(anonymous) @ tfjs@0.6.1:1
o @ tfjs@0.6.1:1
r @ tfjs@0.6.1:1
(anonymous) @ tfjs@0.6.1:1
r @ tfjs@0.6.1:1
(anonymous) @ tfjs@0.6.1:1
o @ tfjs@0.6.1:1
Promise.then (async)
(anonymous) @ app.js:2

Code to reproduce the bug

from keras.applications.xception import Xception
import tensorflowjs as tfjs

model = Xception(weights='imagenet', include_top=False)
tfjs.converters.save_keras_model(model, '/app/data/xception')

tf.loadModel('xception/model.json')
  .then(model => {
    console.log(model);
  })

Issue Analytics

State:
Created 5 years ago
Comments:5

Top GitHub Comments

1reaction

caisqcommented, Apr 4, 2018

There are two issues here:

SeparableConv2D is not implemented yet
The same issue as https://github.com/tensorflow/tfjs/issues/84, related to the padding config key.

I plan to fix 2 soon. I’ll look into 1 later.

0reactions

caisqcommented, Apr 9, 2018

Heads-up: I’m working on separableConv2d in tfjs-core and SeparableConv2D in tfjs-layers

Top Results From Across the Web

Keras Applications

Model Size (MB) Top‑1 Accuracy Top‑5 Accuracy Parameters Depth Time (ms) per infer... Xception 88 79.0% 94.5% 22.9M 81 109.4 VGG16 528 71.3% 90.1% 138.4M...

A Simple Guide to Using Keras Pretrained Models

All pretrained models are available in the application module of Keras. First, we have to import pretrained models as follows.

Transfer Learning in Keras with Computer Vision Models

The pre-trained model can be used as a separate feature extraction program, in which case input can be pre-processed by the model or...

Keras - Applications - Tutorialspoint

Keras applications module is used to provide pre-trained model for deep neural networks. Keras models are used for prediction, feature extraction and fine ......

Module: tf.keras.applications | TensorFlow v2.11.0

Keras Applications are premade architectures with pre-trained weights.