HomeTutorials
[AI knowledge learning]Convolution in One Dimension for Neural Networks
42

[AI knowledge learning]Convolution in One Dimension for Neural Networks

AI  Learning Assistant No 1
AI Learning Assistant No 1
August 17th, 2023

After learning two-dimensional convolution, learning one-dimensional convolution will fall into a misunderstanding, mistakenly thinking that the one-dimensional convolution kernel is convolved on a line. However, in fact, the input of one-dimensional convolution is a vector and a convolution kernel, and the output is also a vector. In this course, you will learn the kernel, equation and backpropagation of one-dimensional convolution.

By Brandon Rohrer

Senior computer scientist, author of Data Science and Robots English website. His articles cover all aspects of computer science and are perfect for the novice to learn, as well as for the experienced to reflect on research or career directions. As of now, his YouTube subscribers have reached 84,500.

[Computer Science]1D convolution for neural networks, part 1: Sliding dot product

[Computer Science]1D convolution for neural networks, part 2: Convolution copies the kernel

[Computer Science]1D convolution for neural networks, part 3: Sliding dot product equations longhand

[Computer Science]1D convolution for neural networks, part 4: Convolution equation

[Computer Science]1D convolution for neural networks, part 5: Backpropagation

[Computer Science]1D convolution for neural networks, part 6: Input gradient

[Computer Science]1D convolution for neural networks, part 7: Weight gradient

[Computer Science]1D convolution for neural networks, part 8: Padding

[Computer Science]1D convolution for neural networks, part 9: Stride

[Computer Science]Implement 1D convolution, part 1: Convolution in Python from scratch

[Computer Science]Implement 1D convolution, part 2: Comparison with NumPy convolution()

[Computer Science]Implement 1D convolution, part 3: Create the convolution block

[Computer Science]Implement 1D convolution, part 4: Initialize the convolution block

[Computer Science]Implement 1D convolution, part 5: Forward and backward pass

[Computer Science]Implement 1D convolution, part 6: Multi-channel, multi-kernel convolutions

[Computer Science]Implement 1D convolution, part 7: Weight gradient and input gradient

[Computer Science]Build a 1D convolutional neural network, part 1: Create a test data set

[Computer Science]Build a 1D convolutional neural network , part 2: Collect the Cottonwood blocks

[Computer Science]Build a 1D convolutional neural network , part 3: Connect the blocks into a network structure

[Computer Science]Build a 1D convolutional neural network, part 4: Training, evaluation, reporting

[Computer Science]Build a 1D convolutional neural network, part 5: One Hot, Flatten, and Logging blocks

[Computer Science]Build a 1D convolutional neural network, part 6: Text summary and loss history

[Computer Science]Build a 1D convolutional neural network, part 7: Evaluate the model

Comments

no dataCoffee time! Feel free to comment