The COLAB NB w/ the MNIST example is so good:https://colab.research.google.com/github/tensorflow/model-optimization/blob/master/tensorflow_model_optimization/g3doc/guide/pruning/pruning_with_keras.ipynb …
You can now easily apply weight pruning to your tf.keras models, to obtain smaller models for inference:https://medium.com/tensorflow/tensorflow-model-optimization-toolkit-pruning-api-42cac9157a6a …
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This will go a long way in explaining how the model works for removing local and global bias.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This is fantastic... will definitely be using this!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Finally there... I am als waiting for an optimized reduced model, relearned based on an existing complex model. A zipped version which might even be more general. The complex model could be used to generate more trainig data, and the original data could be used as test data.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
"Weight, weight, don't tell me!"
-
I'm disappointed your New Yorker profile didn't mention any of your dad jokes
- Show replies
New conversation -
-
-
This method seems to prune based purely on weight magnitude. How does this pruning method perform in terms of maintaining performance while increasing sparsity?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Can this work on any given CNN?
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.