-
# 平台(如果交叉编译请再附上交叉编译目标平台):
-- >>>>>>>>>>>>>
-- MNN BUILD INFO:
-- System: Linux
-- Processor: aarch64
-- Version: 2.8.4
-- Metal: OFF
-- OpenCL: OFF
-- OpenGL: OFF
-- Vulkan: OFF
-- …
-
作者您好,您mobilenet v2的模型训练成功了吗?我用您的程序和您提供的数据训练mobilenet,好像没有成功收敛。测试结果也比vgg16要差很多,而且mobilenet v2模型的前向速度比vgg16的模型好像也没快多少,不知是哪儿弄错了。
-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Feature Description
1. Data collection - Gathered a diverse dataset of images containing children and adults.
…
-
i've Built and installed source code of roi_align_api and rod_align_api.
but while running roi_align.py -> roi_align_api.forward(), outputs are all zeros.
all params are default, i didn't change an…
-
## 🐛 Bug
When trying to optimize a PyTorch model for running on iOS devices with Metal backend
```
traced_detector = torch.jit.trace(kp_detector, image_tensor, strict=False)
optimized_traced_det…
-
将MobileNetV2,width=0.5 的onnx模型,首先经过onnx-simplify简化模型,然后使用onnx2ncnn转换ncnn模型,出现Unsupported reduction axes ! 但模型是能生成的,请问这样会有影响吗?
-
## Description
Hello,
while trying to convert an already quantized ONNX model with Q/DQ nodes to an int8 TensorRT engine, I get the following error:
```
[E] Error[2]: [weightConvertors.cpp::…
-
**请将下面信息填写完整,便于我们快速解决问题,谢谢!**
**问题描述**
请在此处详细的描述报错信息
文档介绍的页面描述的是mobilenetV2,但实际网络是mobileV3。我想知道mobilenetV2在哪里下载。
**更多信息 :**
- 用于部署的推理引擎:
- 为什么需要转换为ONNX格式:
- Paddle2ONNX版本:
- 你的联系方式(Ema…
-
### Checklist
- I have searched related issues but cannot get the expected help.
- I have read related documents and don't know what to do.
### Describe the question you meet
I used the offi…
-
### 请提出你的问题 Please ask your question
问题:
ImportError: cannot import name 'MSRA' from 'paddle.fluid.initializer'
(D:\environmentalSupport\anaconda3\envs\pdenv\lib\site-packages\paddle\fluid\initial…