-
### 请提出你的问题 Please ask your question
版本、环境信息：
1）paddlepaddle-gpu 版本:2.6.0.post116
paddlelite 版本 当前develop分支8795a4d3d v2.13rc v2.10
paddleocr版本 v2.7.1
2）Host 环境：x86_64…
-
For example, I want to change the activation function Swish of EfficientNet to Mish, but the official does not seem to provide the Mish activation function?Should I write Mish function and register it…
-
안녕하세요 강의 잘 듣고 있습니다.
**데이터 과학의 Tensorflow (python)** 듣는 도중 **_네번째 딥러닝 - 신경망의 완성:히든레이어_** 강의에서
> H = tf.keras.layers.Dense(8, activation="swish")(X)
> H = tf.keras.layers.Dense(8, activation="swis…
Sup90 updated
3 years ago
-
There's swish implementation bug when trying to convert model to onnx through ```torch.onnx.export()```.
After trace code, I found that you have already defined ```onnx_export``` argument in Backbo…
-
Swift version 5.0 (swift-5.0-RELEASE)
Target: armv6-unknown-linux-gnueabihf
linker LOG:
```
$ swift build -c release -Xswiftc -g -v
which clang
/usr/local/bin/swiftc -sdk / -g -L /home/p…
-
https://github.com/openppl-public/ppl.kernel.cpu/blob/eda9c78c27d8eacb9a482cd6e2d84739ba0cbcdb/src/ppl/kernel/x86/fp32/hard_swish/hard_swish_fp32_sse.cpp#L74
应该是v_dst3
-
Firstly thank you for your exellent job.
When I try to calculate the FLOPs of MixNet which uses *swish* as non-linearity. And *swish* equals to x*sigmoid(x). But there are no hooks for both sigmoid …
-
1.yolov5 训练一个车牌检测模型，转成fp16用NCNN部署到移动端
python3 export.py --weights /yolo/yolov5/weights/best.pt --include onnx --train
python3 export.py --weights /yolo/yolov5/weights/best.pt --img 320 --include onn…
-
🔥 더 깊은 신경망을 만들기 위해 히든 레이어를 만들어 보자 🔥
`히든 레이어` - input layer와 output layer 사이에 있는 망
`활성 화 함수 중 swich 함수는 요즘 많이 쓰이는 활성화 함수 , 보통 이걸로 다 써도 댄다 `
X -> H -> Y
`#모델의 구조를 만듭니다`
X= tf.keras.layer…
-
我将你modules_ncnns.py 替换了ultralytics\nn\modules.py转换模型与没有修改是一样的:
1.yolo task=detect mode=export model=./last.pt format=onnx simplify=True opset=13
2.onnx2ncnn last.onnx last.param last.bin
yolov8n.pa…