EClemMarq comments

Results 6 comments of


                                            EClemMarq

Support for Conv1D Quantization

Thanks, @choesy! Your workaround worked. I'll elaborate here in case it helps someone else. You can insert Reshape() layers into the model and then replace your Conv1D or MaxPool1D layers...

Activity Regularizer not working with quantization aware training (QAT) TF 2.6

> My problem in version TF2.7. > > ```python > class TNN(tfk.Model): > def __init__(self, input_dim): > super(TNN, self).__init__(name='TNN') > self.fn_dense_input = tfk.layers.Dense(units=256, input_dim=input_dim) > self.fn_dense_hidden = tfk.layers.Dense(units=128) > self.fn_dense_output...

Not all checkpoint values were used.

I see the same behavior with a custom model, TensorRT 7.1.3, and TF 2.3.0. If I build the model before saving, I get an identical message when I load it....

A error when using tfmot.quantization.keras.quantize_model to quantize keras model

This issue seems very similar to #362. There may be more help available in that thread, but it is 3 years old.

Tensorboard frontend does not update when the log directory changes.

I see a similar issue in TensorBoard 2.6.0.

Downloading hparams table should include the Trial ID column

Like @sidphbot, I would use this to find model paths and checkpoints. For example, when I do a hyper-parameter tuning run, I have to go to TensorBoard every time I...