Draft for Pipeline - AAU-Dat/P5-Nonlinear-Dimensionality-Reduction GitHub Wiki
Here follows a draft of the pipeline that we will construct:
-
Load the data from MNIST
- Training data (data + label)
- Test data (data + label)
-
Create the data for MNIST+; perform on original MNIST:
- Blurring
- Rotations
- Scaling
- Sharpening of images
- Randomly remove data points
- Randomly remove pixels from the data points
-
Preprocess the data
- Rescale from 0
1 to 0255 for pixel values
- Rescale from 0
-
Perform dimensionality reduction
- PCA
- LDA
- KPCA
- Isomap
-
Train the model (training data)
- Logistic Regression
- Convolutional Neural Network (keras framework)
-
Test the model (test data)
-
Output the model's results for report
- Metrics
- Plots