Yann POURCENOUX comments

Results 5 comments of


                                            Yann POURCENOUX

PCQAT not working if Conv2D has kernel size 1x1

Can I get an update on this? This is one of your tutorials and 1x1 Convs are one of the most used layers in Deep Learning for computer vision

PCQAT not working if Conv2D has kernel size 1x1

Great! Thanks! Looking forward to hearing from you again 😁

Strange behaviour pruning with small datasets

What if you look at the weights of the model after stripping the pruning with [this](https://www.tensorflow.org/model_optimization/api_docs/python/tfmot/sparsity/keras/strip_pruning)?

loc("model/FPN/FPN/FPN//Concatenate_p_1/concat"): error: 'tfl.concatenation' op quantization parameters violate the same scale constraint: !quant.uniform<i8:f32, 9.9999999999999995E-7> vs. !quant.uniform<i8:f32, 4.9130543629871681E-5:-128>

Hi, thank you for your answer. When I convert using `tf-nightly==2.7.0-dev20210822`, the conversion goes through and gives [this tflite model](https://github.com/tensorflow/model-optimization/files/7030753/model.zip). However, when running the benchmark I encounter the following error:...

Dict-of-tensors custom Metric using MirroredStrategy - tf.identity error?

I have been having this problem for quite some time and I can't find a fix, any updates?