-
Hello,
we want to use the k210 on a larger scale and i am currently testing it among other edge devices.
I need to know the exact layer execution times of a pretrained tensorflow model.
I am t…
-
The MaixPy IDE gives reports: "ValueError: [MAIXPY] kpu: load error:2003, ERR_KMODEL_FORMAT: layer_header.body_size
-
I am trying to create a tcp server from my maix dock m1w, inorder to communicate with my android app(i.e the client). To do so i need to bind my host ip and port address to the socket.
When i run:
…
-
As I said in the title if I try to compile a .kmodel with NNCase v0.2.0 Beta4 it makes freeze the python script in MaixPy. Same problem also with Beta3.
Same model compiled with Beta2 is all OK.
May…
-
Hi,
I was following the instructions from https://github.com/kendryte/tensorflow-workspace/commit/d80f0d2ac23287aaf8b116c9cd3769b90c720bee#diff-7d9fc065a89438c0ca7e01e4611d0f3aR30 to generate C cod…
-
Hello, I have an issue where my Dedicated GPU Memory, keeps filling up to almost max, about 23 out of 24 gb vram.
It doesn't seem to matter what I put the gpu weight at, whether it be at 23 gb, 16gb,…
-
uploading a kmodel with uPyloader is capped at 115200, this is very slow and takes many minutes for a small 2.4MB model.
Is there an alternative way to upload a model to the sd card that's fast, and …
-
Dear developer,
I tried to repeat your example code and meet some errors.
1), the N1-rekindle-data-preprocessing.py cannot find the file in the 'models' folder. I solved this problem by downloading…
ghost updated
2 years ago
-
For some reason dev fp8 works better than nf4 when I use lora, Nf4 just use a lot of vram and ram making the generation speed absurdly slow.
With dev fp8 I'm getting speeds like 1.01 s/it and with …
-
(rtx 4060)
ok i tryed booth
with and without with different sampler and schedulers
(resolution 1024x768)
if i generate an image with a lora speed is ~ 3s/it
if i choose same conditions without th…