-
Hello Guys,
Would you add UXGA support as base resolution for windowing mode, and also for taking JPEG images that then to be saved as JPEG image file or added as frames in AVI?
OV2640 can make JPE…
-
**问题**
使用例程中的“编译uint8量化tflite模型”，模型使用的样例中的‘examples/20classes_yolo/model/20classes_yolo.tflite’，修改量化类型为int16，输出模型大小无任何变化，Netron查看ir_target_dep_after_quant下的pb文件，卷积核等各项参数仍然是uint8类型
compile_options.qu…
-
We can't useing the Native code emitter & Viper code emitter on canmv boards! why?
a sample here:
```python
@micropython.viper
def foo(self, arg: int) -> int:
# code
```
```python
@mic…
-
**Describe the bug**
在原始模型中输出中添加了Sigmoid 但是转换成kmodel 之后用Simulator 进行模拟 但是输出的结果都是大于1的
**To Reproduce**
Command line or scripts to reproduce the behavior:
**Expected behavior**
希望能符合我的原始输出
…
-
Describe the bug
用c语言运行模型时报错
报错信息如下
terminate called without an active exception
SYSCALL: sys_exit called by core 0 with 0x1
To Reproduce
用于跑声音异常检测的kmodel，在加载过程没有报错,运行过程没有反应
Expected behavi…
-
在aa09f2b中的bsp里初始化到装载sd文件系统时出现：
> initialize mnt_initset spi freq 400198
Function[rt_mutex_take] shall not be used in ISR
(0) assertion failed at function:rt_mutex_take, line number:921
将**drv…
-
**Is your feature request related to a problem? Please describe.**
I want to try running the yolox kmodel on k210 with micro python. The repo has C/C++ code in nncase-master\examples\yolox but there …
-
**Describe the bug**
我用nncase1.8.x 生成kmodel模型 total : 3.15MB，然后使用从flash读取的方式读取模型读取模型会直接提示这个错误。
我用nncase1.7.x 生成kmodel模型 total : 3.15MB，然后使用从flash读取的方式读取模型读取模型不会提示这个错误，但是运行一轮识别之后就还是会出现这个错误，我尝试去除我的后期处…
-
I am trying to use a modified version of the MobileNetV2 model of TensorFlow Keras on the Kendryte k210, for simple image recognition.
The model has been modified so that it (hopfully) meets all the …
-
The yolox_nano_224.onnx model present at https://github.com/kendryte/nncase/tree/master/examples/yolox/model
doesn't show correct predictions for an image.
I setup the yolox repo from https://gith…