PyTorch torch.nn.Conv3d Function
PyTorch torch.nn Reference Manual
torch.nn.Conv3dIs a 3D convolution module in PyTorch.
It processes 3D inputs, such as videos or medical images (depth x height x width).
Function Definition
torch.nn.Conv3d(in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, bias=True)
Input Shape
(batch, channels, depth, height, width)
Usage Examples
Example 1: Basic Usage
Example
import torch
import torch.nn as nn
# 3D convolution
conv3d = nn.Conv3d(in_channels=3, out_channels=32, kernel_size=3, padding=1)
# Input: batch=2, channels=3, depth=16, height=32, width=32
x = torch.randn(2, 3, 16, 32, 32)
output = conv3d(x)
print("Input shape:", x.shape)
print("Output shape:", output.shape)
import torch.nn as nn
# 3D convolution
conv3d = nn.Conv3d(in_channels=3, out_channels=32, kernel_size=3, padding=1)
# Input: batch=2, channels=3, depth=16, height=32, width=32
x = torch.randn(2, 3, 16, 32, 32)
output = conv3d(x)
print("Input shape:", x.shape)
print("Output shape:", output.shape)
Example 2: Video Classification
Example
import torch
import torch.nn as nn
class VideoCNN(nn.Module):
def __init__(self, num_classes=10):
super(VideoCNN, self).__init__()
self.conv1 = nn.Conv3d(3, 32, kernel_size=3, padding=1)
self.pool = nn.MaxPool3d(2)
self.conv2 = nn.Conv3d(32, 64, kernel_size=3, padding=1)
self.gap = nn.AdaptiveAvgPool3d(1)
self.fc = nn.Linear(64, num_classes)
def forward(self, x):
x = torch.relu(self.conv1(x))
x = self.pool(x)
x = torch.relu(self.conv2(x))
x = self.gap(x)
x = x.view(x.size(0), -1)
return self.fc(x)
model = VideoCNN()
x = torch.randn(4, 3, 16, 112, 112) # 16-frame video
output = model(x)
print("Video:", x.shape, "-> Class:", output.shape)
import torch.nn as nn
class VideoCNN(nn.Module):
def __init__(self, num_classes=10):
super(VideoCNN, self).__init__()
self.conv1 = nn.Conv3d(3, 32, kernel_size=3, padding=1)
self.pool = nn.MaxPool3d(2)
self.conv2 = nn.Conv3d(32, 64, kernel_size=3, padding=1)
self.gap = nn.AdaptiveAvgPool3d(1)
self.fc = nn.Linear(64, num_classes)
def forward(self, x):
x = torch.relu(self.conv1(x))
x = self.pool(x)
x = torch.relu(self.conv2(x))
x = self.gap(x)
x = x.view(x.size(0), -1)
return self.fc(x)
model = VideoCNN()
x = torch.randn(4, 3, 16, 112, 112) # 16-frame video
output = model(x)
print("Video:", x.shape, "-> Class:", output.shape)
Example 3: Parameter Calculation
Example
import torch
import torch.nn as nn
conv = nn.Conv3d(64, 128, kernel_size=3, padding=1)
# Parameter count calculation: out_ch * in_ch * k_d * k_h * k_w + bias
params = 128 * 64 * 3 * 3 * 3
print("Number of parameters:", params)
print("Weight shape:", conv.weight.shape)
import torch.nn as nn
conv = nn.Conv3d(64, 128, kernel_size=3, padding=1)
# Parameter count calculation: out_ch * in_ch * k_d * k_h * k_w + bias
params = 128 * 64 * 3 * 3 * 3
print("Number of parameters:", params)
print("Weight shape:", conv.weight.shape)
Use Cases
- Video processing: action recognition
- Medical imaging: CT、MRI
- 3D segmentation: voxel data
Note: 3D convolution requires a large amount of computation and needs more GPU memory.
Other Extensions