conv_net.py

import torch
import torch.nn as nn
import torch.nn.functional as F


class ConvNet(nn.Module):
    def __init__(self):
        super(ConvNet, self).__init__()
        self.conv1 = nn.Conv2d(3, 16, 3, 1, 1)
        self.bn1 = nn.BatchNorm2d(16)
        self.conv2 = nn.Conv2d(16, 32, 3, 1, 1)
        self.bn2 = nn.BatchNorm2d(32)
        self.conv3 = nn.Conv2d(32, 64, 3, 1, 1)
        self.bn3 = nn.BatchNorm2d(64)
        self.dropout = nn.Dropout2d(0.1)
        self.fc = nn.Linear(20 * 20 * 64, 16)
        
    def forward(self, x):
        # HWC to CHW
        x = x.permute(0, 3, 1, 2)
        
        x = F.relu(self.bn1(self.conv1(x)))
        x = F.relu(self.bn2(self.conv2(x)))
        x = F.relu(self.bn3(self.conv3(x)))
        x = F.max_pool2d(x, 2)
        x = self.dropout(x)
        x = torch.flatten(x, 1)
        x = self.fc(x)
        output = F.log_softmax(x, dim = 1)
        return output