https://github.com/anindya-prithvi/mnist_caelatentspace

Convolutional AutoEncoders (MNIST) and their generative capabilities (kind of amazing)
https://github.com/anindya-prithvi/mnist_caelatentspace

autoencoder-mnist linear-discriminant-analysis machine-learning mnist onnx onnxruntime onnxruntime-web plt python

Last synced: 4 months ago
JSON representation

Convolutional AutoEncoders (MNIST) and their generative capabilities (kind of amazing)

• Host: GitHub
• URL: https://github.com/anindya-prithvi/mnist_caelatentspace
• Owner: Anindya-Prithvi
• Created: 2022-05-20T13:05:19.000Z (about 3 years ago)
• Default Branch: main
• Last Pushed: 2022-05-23T17:04:13.000Z (about 3 years ago)
• Last Synced: 2025-01-16T05:12:30.296Z (5 months ago)
• Topics: autoencoder-mnist, linear-discriminant-analysis, machine-learning, mnist, onnx, onnxruntime, onnxruntime-web, plt, python
• Language: Jupyter Notebook
• Homepage: https://anindya-prithvi.github.io/MNIST_CAELatentSpace/
• Size: 6.92 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Objective

Observe the Latent space of Convolutional AutoEncoder for a classification dataset (MNIST)

# Architecture of Auto Encoder

The encoder consists of multiple convolutions and batchnormalizations only. We scale a 28x28 to 7x1 vector.  

The decoder is again transposed convolutions and batchnormalizations, i.e. 7x1 to 28x28.  

Optimizers/LR/decays used have not been hyperparameter optimized yet. Therefore the current loss is 0.02 for 50 epochs.  

![CAE Output](https://github.com/Anindya-Prithvi/MNIST_CNNLatentSpace/blob/main/assets/CAEresult.png)

# To Observable

We encode the training images to our latent space as usual. However we then apply LDA transformations with `n_components=2`. Plotting the transformed data we could observe well formed clusters (refer notebook). Also, there were obviously some overlapping ones, to resolve them, a 3D LDA transformation was obtained. The clusters were amazingly distinguishable with almost 0 overlap.  

## 2D LDA result.  

![2D LDA result](https://github.com/Anindya-Prithvi/MNIST_CNNLatentSpace/blob/main/assets/LDA2.png)

## 3D LDA plot © [Aflah](https://github.com/aflah02/)  

[3D LDA result - Interactive (Click)](https://anindya-prithvi.github.io/filehost/plotlypage.html)  

![image](https://user-images.githubusercontent.com/29653551/169578739-40f9c343-d80b-4479-855b-9a4d4062adc5.png)

# Observations

1. We can generate multiple fake samples given the latent space encoding distributed on multivariate gaussian distribution with parameters from MLE estimates of the training data (of a specific class) in latent space. 

2. Latent space was initally made keeping in mind a 7-segment decoder. But guess what (HINT: Look at the means)

3. All of the following is synthetic data  

   ![synthesis](https://github.com/Anindya-Prithvi/MNIST_CNNLatentSpace/blob/main/assets/synthetic.png)  

   ```py

   array([

       3, 2, 2, 2, 7, 6, 6, 2, 2, 2, 2, 1, 2, 6, 2, 2, 

       2, 2, 6, 2, 2, 2, 2, 2, 4, 2, 2, 2, 2, 2, 1, 2, 

       8, 2, 2, 2, 2, 6, 2, 2, 6, 0, 7, 2, 0, 2, 2, 9, 

       5, 2, 2, 2, 5, 2, 2, 2, 6, 6, 2, 2, 2, 2, 2, 2, 

       2, 2, 2, 2, 4, 8, 2, 2, 2, 2, 3, 2, 2, 2, 2, 2, 

       6, 2, 8, 2, 2, 2, 6, 2, 2, 2, 2, 2, 2, 8, 3, 2, 

       2, 2, 6, 2, 8, 2, 2, 2, 2, 2, 2, 5, 2, 2, 2, 2, 

       2, 6, 2, 2, 2, 2, 2, 2, 2, 8, 2, 6, 8, 2, 2, 2], dtype=int64)

4. A demo is now live at [ghpages](https://anindya-prithvi.github.io/MNIST_CAELatentSpace/). Drag the sliders to move around the Latent space.