I made available VGG16, VGG19, ResNet50: Keras code +ImageNet weights for both TF and Theano.https://github.com/fchollet/deep-learning-models …
-
the reason I ask is because I'm working on tutorials to train ImageNet-level networks using Python + Keras.
in a multi-replica setup? I've been training lots of Keras ImageNet models on a bunch of GPUS (TF) so I could give you tips
-
-
That would be awesome, I would really appreciate it. I have 4 Titan X GPUs that I am using for training now.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
is scoping possible in keras? i.e. with framework.arg_scope([tflearn.variables.variable], device='/cpu:0'): blah ?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I'm also guessing that for grad computations & averaging you directly use the tf.
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.