Elena Zhelezina comments

Results 7 comments of


                                            Elena Zhelezina

[clustering] Possible wrong implementation of get_weight_from_layer

Hi @tisma, The function `get_clusterable_weights` in your implementation does not return what is expected by the clustering algorithm. It should be ``` def get_clusterable_weights(self): return [('kernel', self.kernel)] ``` We have...

[clustering] Possible wrong implementation of get_weight_from_layer

Hi @tisma To pass to the clustering algorithm what should be clustered in your layer, you need to take a look at attributes with weights. This is an advanced usage,...

PCQAT not working if Conv2D has kernel size 1x1

Hi @rino20 Yes, we will take a look today/tomorrow at this issue.

[clustering] Possible wrong call of the centroids initializer from the ClusterWeights wrapper

Hi @aserenko Thank you for the reported bug. The fix is in this PR https://github.com/tensorflow/model-optimization/pull/951

Not able to cluster Conv1DTranspose layer

Hi @rameshkunasi Thank you for the reporting the problem. This PR https://github.com/tensorflow/model-optimization/pull/952 adds the requested support of Conv1DTranspose. I added unit test based on your example - please take a...

RFC: Mechanism to indicate that tflite model is structurally pruned.

Yes, when this flag is set, we need to check every weight(tensor) of the model. As result, the conversion time will be bigger, but inference time will be better. Usually,...

RFC: Mechanism to indicate that tflite model is structurally pruned.

closing as it is not relevant anymore