A Tiny CNN by Hand: Conv → ReLU → Pool → Score
This lesson is part of the course
In this lesson
You push a 6×6 handwritten "1" through a complete mini CNN: a vertical-line kernel with bias, ReLU, 2×2 max pooling and a dense layer with a sigmoid, and read the network's probability that the image shows a 1.
Buy the course once for $9.99 to unlock every lesson and lab, the final exam and the certificate – lifetime access, no subscription.
Lessons in this course
- Pixels as Numbers: RGB to Greyscale with Weighted Luminance Locked
- Brightness, Contrast & the Histogram (FREQUENCY, with a Chart) Locked
- Thresholding & Binary Masks: Separating an Object from the Background Locked
- 3×3 Convolution with SUMPRODUCT: a Gaussian Blur Locked
- Sharpening & Edge Detection: Sobel Gx, Gy and Gradient Magnitude Locked
- Padding & Stride: the Output-Size Formula and a Stride-2 Convolution Locked
- Max & Average Pooling: Shrinking Feature Maps Locked
- A Tiny CNN by Hand: Conv → ReLU → Pool → Score Locked