Add pretrained model weights for Keras
See original GitHub issueKeras has a keras.applications module with pretrained models and automatically downloads their weights. Since the Kaggle Kernels have no external internet access, these cannot be downloaded. It would however work to add these weights to the docker-image so they were always available (since keras first checks a local cache before downloading, https://github.com/fchollet/keras/blob/master/keras/applications/vgg16.py#L162) https://github.com/fchollet/keras/blob/master/keras/utils/data_utils.py#L117). Particularly as a number of the tasks from the YouTube challenge to the Cervical Cancer contest would benefit from using pretrained models to create feature vectors
Issue Analytics
- State:
- Created 6 years ago
- Comments:7 (4 by maintainers)
Top Results From Across the Web
A Simple Guide to Using Keras Pretrained Models
To use the pretrained weights we have to set the argument weights to imagenet . The default value is also set to imagenet...
Read more >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...
Read more >Save and load Keras models | TensorFlow Core
Saving the weights values only. This is generally used when training the model. Let's take a look at each of these options. When...
Read more >How can I use pre-trained weights from an existing CNN ...
You can train this model in the same way you trained your previous CNN. Keras applications provides access to many models such as...
Read more >Transfer Learning in Keras with Computer Vision Models
The first time a pre-trained model is loaded, Keras will download the required model weights, which may take some time given the speed...
Read more >
Top Related Medium Post
No results found
Top Related StackOverflow Question
No results found
Troubleshoot Live Code
Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free
Top Related Reddit Thread
No results found
Top Related Hackernoon Post
No results found
Top Related Tweet
No results found
Top Related Dev.to Post
No results found
Top Related Hashnode Post
No results found

I uploaded the model files as a data set here: https://www.kaggle.com/ekkus93/keras-models
Here’s an example how you can use it: https://www.kaggle.com/ekkus93/keras-models-test-v2
For those who wanted to use other models except VGG16 or VGG19 can use the following dataset, https://www.kaggle.com/gaborfodor/keras-pretrained-models Thought I should add it as an extension.