-
Hello,
Is there an "ncc" Windows application available to convert the models trained with Tensorflow 2.x (ultimately converted to TFLite with MLIR support) to convert TFLite models to kmodel? I am tr…
-
### Help
https://github.com/kendryte/k230_canmv/blob/main/tests/ai_demo/person_kp_detect.py
请问上面这个demo中import的aidemo库文件我可以在哪里查看到
-
**Describe the bug**
在我使用simulator进行验证的时候，它输出了nan。我使用yolov8的onnx以及kmodel去验证我的kmodel_input.bin和onnx_input.bin余弦相似度为0.99，但是一切换到yolov5的onnx和kmodel，他就变成nan了
**Origin model and code**
这是生成kmodel的代码
`…
-
Using "MicroPython v0.5.0-200-g8f6aa7b on 2020-08-31; Sipeed_M1 with kendryte-k210" on board: Maixduino
There is no status function for NIC created as ESP32_SPI that seems a bit strange as the func…
-
**Describe the bug**
将ConvNext预训练模型转kmodel时，nncase 编译器输出："2.Optimize target independent..." 后退出。无任何其他提示
**To Reproduce**
预训练模型可以在以下地址获取：
https://dl.fbaipublicfiles.com/convnext/convnext_tiny_1…
-
**Describe the bug**
A clear and concise description of what the bug is.
I am working on a transformer model in pytorch. After cleaning it up, I converted to onnx format and pass the onnx model to h…
-
https://github.com/kendryte/kendryte-standalone-sdk/issues/122
-
When using the command "bazel build", it came out the error with this tip "Python Configuration Error: Cannot find python in PATH, please make sure python is installed and add its directory in PATH, o…
-
### Other
Hello,
Could you provide list of compatible RPI Cameras?
-
### Help
When NuttX running on big core, test app `getprime` takes about 11158ms, while it took only 308ms on little core.
I guess some settings for big core is missing (e.g. cache) here, can …
yf13 updated
10 months ago