https://github.com/githubharald/analyze_pooling

https://github.com/githubharald/analyze_pooling

Last synced: 3 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/githubharald/analyze_pooling
• Owner: githubharald
• License: mit
• Created: 2022-07-21T19:29:58.000Z (almost 3 years ago)
• Default Branch: master
• Last Pushed: 2022-07-21T19:40:04.000Z (almost 3 years ago)
• Last Synced: 2025-01-20T06:13:48.863Z (5 months ago)
• Language: Python
• Homepage: https://harald-scheidl.medium.com/d2f5a7866135
• Size: 141 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

        
# Analyze pooling layers

This code allows **analyzing** the **effect** of the **pooling layers** in a **CNN**.

It implements a simple CNN model trained to distinguish between filled rectangles (class 0) and circles (class 1).

The model takes a 224x224 image as input and outputs a value between 0 and 1.

![img](doc/inference.png)

Pooling layers cause the model to be shift variant: 

shifting the input image by 1px causes the output to be different, even though CNNs are often said to be shift invariant.

## Getting started

Analyze the effect of the single pooling layer in the pre-trained network:

* Go to the folder `src/`

* Then, execute ```python main.py --mode analyze --num_pooling_layers 1```.

This command shifts an input shape (rectangle, class 0) to be classified all around the input image and then records the prediction (should be close to 0 because of predicting class 0) for all the tested image locations. The output should look as follows:

![img](doc/analysis.png)

As can be seen, depending on the position of the shape in the input image the output slightly differs. This shows that CNNs with pooling layers are not shift invariant.

## Usage

General usage:

```python main.py --mode {train,infer,analyze} --num_pooling_layers {0,1,2}```.

Parameters:

* mode: first, train the model, then, you can either infer a few samples or analyze the effect of the pooling layers

* num_pooling_layers: number of pooling layers in the model, either 0, 1 or 2

Example to train the model:

```python main.py --mode train --num_pooling_layers 1```.

Example to infer a few samples with the trained the model:

```python main.py --mode infer --num_pooling_layers 1```.

## More information

This [article](https://harald-scheidl.medium.com/d2f5a7866135) discusses the effect and shows suggestions on how to reduce it.