You can now literally do: ``` model = VGG16(weights='imagenet') preds = model.predict(imgs) print(decode_predictions(preds)) ```
I made available VGG16, VGG19, ResNet50: Keras code +ImageNet weights for both TF and Theano.https://github.com/fchollet/deep-learning-models …
-
-
-
amazing
End of conversation
New conversation -
-
-
Nice. What solution did you do for model distribution? I've wondered the same thing - both for sharing trained models and datasets
-
Using Github file release system currently.
End of conversation
New conversation -
-
-
Awesome job! Super excited about this.
-
Were these models trained from scratch? Or the weights imported/converted from Caffe models?
- Show replies
New conversation -
-
-
Thank you! Where the ResNet50 weights ported or trained? 200 layer model exists here: https://github.com/facebook/fb.resnet.torch …
-
@github weights were ported from Caffe.
End of conversation
New conversation -
-
-
Excellent, thank you!!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Merci! Now we only need a ROIPooling layer in Keras to have everything a CV person could desire to do kickass object detection!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.