Simon Narduzzi comments

Results 9 comments of


                                            Simon Narduzzi

How to use default n bit in QAT ?

Hello, any update on the default n bit quantization? Is a tutorial available by any chance?

kmodel v5 support

Any update on this?

optimize IDE uploading speed for big files

I'm facing the same issue with MaixPy v0.3.2

tflite2kmodel : Error (Dimensions must be equal)

Hi @Yhaw, I think I ran into the same problem and solved it by using convolutions with "same" padding. Furthermore, check that each activation map does not exceed 2MB, as...

tflite2kmodel : Error (Dimensions must be equal)

Hi @Yhaw, You have to modify the model to make sure every convolutional layer is using "same" as padding. One way to do it is to rewrite the model from...

tflite2kmodel : Error (Dimensions must be equal)

By the way, I have just opened [this issue](https://github.com/kendryte/canmv/issues/35#issue-1721728061) about how to use the latest version of NNCase with the Maix M1 Dock. Hopefully it will solve your problem as...

Code stops working after few seconds on Maix M1 Dock

The error seems to originate from the IDE as the code continues to run on the chip.

Code stops working after few seconds on Maix M1 Dock

Hi @kendryte747, I have updated my IDE to the latest release (Commit: `v2.9.2-2-0-g132467a`), but the bug is not solved. Is it originating from the editor of from the firmware?

Activity Regularizer not working with quantization aware training (QAT)

It seems that the losses are a list of several losses. Can you try the following code instead? ```python print([x.numpy() for x in quantized_model.losses]) ```