---
title: Conv
description: "Conv2d：常用于边缘检测、特征提取。"
url: https://www.hikunpeng.com/document/detail/zh/kunpenghpcs/hpckit/devg/KunpengHPCKit_developer_205.html
sourcePath: /source/zh/kunpenghpcs/hpckit/devg/KunpengHPCKit_developer_205.html
indexId: 8941a01fdf7f4957b0c2bb62350028fcddb155765d8c4069a57acb8dd7574dc767
---
# Conv

#### 场景说明

Conv2d：常用于边缘检测、特征提取。

Conv3d：常用于时空特征提取。

目前KuDNN支持torch.float16和torch.float32数据类型，其他数据类型会使用开源分支。


#### 示例代码

```
import torch
import torch.nn as nn

#使能KuDNN
torch._C._set_kdnn_enabled(True)

# Conv2d示例
conv2d = nn.Conv2d(in_channels=3, out_channels=16, kernel_size=3, stride=1)
x = torch.randn(8, 3, 32, 32)  # [batch, channel, H, W] # 默认为fp32类型
y = conv2d(x)  # 输出 [8, 16, 30, 30]
print("Conv2d输出形状", y.shape)
print(y)

# Conv3d示例
conv3d = nn.Conv3d(1, 8, kernel_size=(3,3,3)) 
x = torch.randn(4, 1, 10, 64, 64)  # [batch, channel, depth, H, W] # 默认为fp32类型
y = conv3d(x)  # 输出 [4, 8, 8, 62, 62]

print("Conv3d输出形状", y.shape)
print(y)
```