-
When I convert it to the Pytorch version, I have the following problems. Can you help me solve them?thanks
error:
x_gap = x.mean(axis=2,keepdims=True).mean(axis=3,keepdims=True)
TypeError: m…
-
![image](https://user-images.githubusercontent.com/53890383/90268080-068cd480-de89-11ea-81d4-73f681a4c315.png)
我吐了,为什么MegStudio中加载模型,数据集都巨慢?
向2G致敬???
-
## 环境
1.ubuntu :
2.0.3:
3.python3.7.3:
4.resnet:
## 复现步骤
1. cd official/vision/classification/resnet
2. python train.py --data=/path/to/imagenet --ngpus=8 --workers=8
## 请提供完整的日志及报错信息
P…
-
## 请简要描述您的需求
More than one file was found with OS independent path 'lib/arm64-v8a/libshufflenet_inference.so'. If you are using jniLibs and CMake IMPORTED targets, see https://developer.android.com/s…
-
## 请简要描述您的需求
浏览API应该是用的数据并行的方式,之后有支持模型并行的计划吗?
-
## 环境
1.系统环境:Ubuntu16.04
2.MegEngine版本:0.3.1
3.python版本:3.6
你好,我在用 MegEngine 写代码的时候遇到了一个 Bug。跑下面这段代码时,显存不断增加,感觉木有释放。Bug 代码如下:
```python3
import megengine as mge
import megengine.module as…
amusi updated
4 years ago
-
Tensor clamp
## 背景
I am doing a project regarding object detection. So I need to clamp the predicted boxes. Does the tensor in the MegEngine have clamp methods like tensors in the PyTorch framew…
-
## 环境
1.系统环境及版本:(例 Ubuntu 18.04)
2.MegEngine版本:(若您是自行编译,请提供 fork 时对应 commit id 、编辑器及版本)
3.python版本:
4.CPU/GPU :(若您使用GPU进行训练,请提供您的CUDA和cuDNN版本号)
5.Git 版本:
6.GCC版本:
## 复现步骤
1.
2.
3.
## …
-
Project
megengine :https://pypi.org/project/MegEngine/
Size of release
850M
Which indexes
both
Reasons for the request
megengine is deep learning framework,the installation package contai…
-
## 请简要描述您的需求
![image](https://user-images.githubusercontent.com/11574772/77511115-543d7c00-6eab-11ea-8156-71eb2f6fce5a.png)