-
Hi sir very nice work I appreciate it . I just want to know what has not been implemented yet ?
-
I have done some testing with the latest Vulkan drivers on a Raspberry Pi 4 (64-OS). Knowing the driver is still under construction, the results were a great disappointment. No acceleration at all, it…
-
您好,我在利用benchncnn测试模型遇到一些速度问题
测试平台: linux服务器、骁龙845芯片
ncn版本: 20200727
编译链接:https://github.com/Tencent/ncnn/wiki/how-to-build#build-for-linux-x86
https://github.com/Tencent/ncnn/wiki/how-to-build#b…
-
I see the current version only allows `ssdMobilenetv1` but seems like `v2` is already available.
How hard is to get to use the newer version instead? Also, what's needed to _make it work_ on this …
-
`-------------------------------------------------------------------------------------
PaddleFace
+----------------+---…
-
```
pi@raspberrypi:~/ncnn/build/benchmark $ ./benchncnn 4 4 0 -1 1
loop_count = 4
num_threads = 4
powersave = 0
gpu_device = -1
cooling_down = 1
squeezenet min = 179.50 max = 183.…
nihui updated
12 months ago
-
The output layer for **conf** prediction is of shape _**(None, 896, 1)**_ but my input label shape is **_(BATCH_SIZE, 1)_**.
I am getting a shape mismatch error here.
`Traceback (most recent call…
-
Hello,
While trying to include and try Bert on a website (code attached), I have this error (attached) I do not know if the error comes from an inconsistency in my code or an error from you (as it …
Aub16 updated
10 months ago
-
### Weekly-Review
- 继续对比人脸检测方法Retinaface
1. Retinaface核心思想:利用4多任务学习策略分别预测人脸分数、人脸框、五点关键点以及self-supervised解码分支预测人脸三维位置
1. retinaface作者提供的mxnet版本与第三方pytorch实现的版本在检测精度上几乎无差别,而且检测率也非常高,与真值人脸数量等同,对…
-
We often process images in the following way before feeding them to models:
`tf.mul(tf.sub(inputImage.toFloat().div(255), 0.5), 2)`
This could be achieved with a single kernel, following TensorF…