PyTorch torch.nn.Conv3d 関数
torch.nn.Conv3dPyTorch の3次元畳み込みモジュールです。
3次元入力（深さ x 高さ x 幅）を処理します。例えば、動画や医用画像などです。
関数定義
torch.nn.Conv3d(in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, bias=True)
入力形状
(batch, channels, depth, height, width)
使用例
例1: 基本的な使い方
実例
import torch
import torch.nn as nn
# 3D 畳み込み
conv3d = nn.Conv3d(in_channels=3, out_channels=32, kernel_size=3, padding=1)
# 入力: batch=2, チャンネル=3, 深さ=16, 高さ=32, 幅=32
x = torch.randn(2, 3, 16, 32, 32)
output = conv3d(x)
print("入力形状:", x.shape)
print("出力形状:", output.shape)
import torch.nn as nn
# 3D 畳み込み
conv3d = nn.Conv3d(in_channels=3, out_channels=32, kernel_size=3, padding=1)
# 入力: batch=2, チャンネル=3, 深さ=16, 高さ=32, 幅=32
x = torch.randn(2, 3, 16, 32, 32)
output = conv3d(x)
print("入力形状:", x.shape)
print("出力形状:", output.shape)
例2：動画分類
実例
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フレームの動画
output = model(x)
print("動画:", x.shape, "-> カテゴリ:", 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フレームの動画
output = model(x)
print("動画:", x.shape, "-> カテゴリ:", output.shape)
例3: パラメータ計算
実例
import torch
import torch.nn as nn
conv = nn.Conv3d(64, 128, kernel_size=3, padding=1)
# パラメータ数計算: out_ch * in_ch * k_d * k_h * k_w + bias
params = 128 * 64 * 3 * 3 * 3
print("パラメータ数:", params)
print("重み形状:", conv.weight.shape)
import torch.nn as nn
conv = nn.Conv3d(64, 128, kernel_size=3, padding=1)
# パラメータ数計算: out_ch * in_ch * k_d * k_h * k_w + bias
params = 128 * 64 * 3 * 3 * 3
print("パラメータ数:", params)
print("重み形状:", conv.weight.shape)
使用シーン
- 動画処理: アクション認識
- 医用画像: CT、MRI
- 3D セグメンテーション: ボクセルデータ
注意: 3D 畳み込みは計算量が多く、多くのVRAMが必要です。
その他の拡張