Adding a new convolutional layer

Your project lead provided you with a new CNN model. Let's take a look at the model's architecture and append a new convolutional layer to it.

The model is available as CNNModel. The packages torch and torch.nn as nn have been imported.

This exercise is part of the course

Deep Learning for Images with PyTorch

View Course

Exercise instructions

Instantiate a model from the CNNModel class and access the convolutional layers.
Create a new convolutional layer with in_channels equal to existing layer's out_channels, out_channels set to 32, and stride and padding both set to 1, and a kernel_size of 3; assign it to conv2.
Append the new layer to the model, calling it "conv2".

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

# Create a model
model = ____
print("Original model: ", model)

# Create a new convolutional layer
conv2 = ____

# Append the new layer to the model
model.____(____)
print("Extended model: ", model)

Edit and Run Code