ProgLearn
Using ProgLearn
Overview
The Library
The Learners
Wrap Up
Install
Install from PyPi
Install from Github
Python package dependencies
Hardware requirements
OS Requirements
Testing
Tutorials
Installation Tutorial
1. Installation
2: Package Setup
Estimation of Bayes’ Error for Gaussian Parity
What is Bayes’ Error?
Problem Definition: Gaussian Parity
Estimation of Bayes’ Error (Analytically)
Estimation of Bayes’ Error (Numerically)
How to Run UncertaintyForest
Import required packages and set parameters for the forest
Create and train our UncertaintyForest
Produce a metric of accuracy for our learner
What’s next?
Analyzing the UncertaintyForest Class by Reproducing Posterior Estimates
Import Required Packages
Specify Parameters
Specify Learners
Generate predicted posteriors
Create Figure 1
Analyzing the UncertaintyForest Class by Reproducing Conditional Entropy Estimates
Import Required Packages
Specify Parameters
Specify Learners
Plot Figure 2
Analyzing the UncertaintyForest Class by Reproducing Mutual Information Estimates
Import Required Packages
Specify Parameters
Specify Learners
Plot Figure 3
Scene Segmentation with Random Forests
I. Preprocessing of Images
II. Scene Segmentation: Scikit-Image
Proglearn: Scene Segmentation of ISIC using Scikit-Image
0. Environment Setup
I. Preprocessing of Images
II. Scene Segmentation using Scikit
III. Conclusion
Scene Segmentation with Neural Networks
Preprocessing
Neural Network
Experiments
FTE/BTE Experiment for food-101
Import necessary packages and modules
Experimental Set-up
food-101 Data Generation
Train the model and perform validation
Calculating FTE, BTE, TE, and Accuracy
Plotting FTE, BTE, TE, and Accuracy
FTE/BTE Experiment for MNIST & Fashion-MNIST
Benchmark Individual Datasets
FTE/BTE Between Datasets
Label Shuffle Experiment
Import necessary packages and modules
Load CIFAR100 data
Define hyperparameters for the model and preprocess data
Train the model and perform validation
Function to calculate backward transfer efficiency
Plotting the backward transfer efficiency
Random Classification Experiment
Synergistic Learning
Choosing hyperparameters
Loading datasets
Running experiment
Plotting backward transfer efficiency
Rotation CIFAR Experiment
Hyperparameters
Algorithms
Experiment
Rotation CIFAR Plot
Expected Results
FAQs
Recruitment Experiment
Recruitment Within Datasets: CIFAR10x10
Recruitment Between Datasets: MNIST/Fashion-MNIST
Recruitment Across Datasets
FTE/BTE Experiment
Recruitment Experiment
Other Experiments
Spiral Experiment
Function Import
Data Creation
Training/Experiment
Results
Spoken Digit Experiment
Import necessary packages and modules
Load spoken_digit data and extract features
Inspect data
Run Synergistic Learning
Calculate and plot transfer efficiency
Shuffled speaker
Gaussian XOR and Gaussian R-XOR Experiment
Classification Problem
The Experiment
Visualizing the Results
Various Angles vs BTE
Number of training samples vs BTE
Gaussian XOR and Gaussian R-XOR BTE with CPD Experiment
Classification Problem with Domain Adaptation
The Experiment (Various Angles vs BTE)
Gaussian XOR and Gaussian R-XOR BTE with Supervised Alignment
Classification Problem with Domain Adaptation
The Experiment (Various Angles vs BTE)
Gaussian XOR and Gaussian R-XOR Experiment with Task Unaware Settings
Ksample test
Task aware BTE and generalization error (XOR)
Task Unaware: K-sample testing “dcorr”
Gaussian XOR and Gaussian XNOR Experiment
Classification Problem
The Experiment
Visualizing the Results
Double-descent phenomena on Random Forest
Discussion
API Reference
Transformers
NeuralClassificationTransformer
TreeClassificationTransformer
Voters
TreeClassificationVoter
KNNClassificationVoter
Deciders
SimpleArgmaxAverage
Progressive_Learner
ClassificationProgressiveLearner
Forest
LifelongClassificationForest
Network
LifelongClassificationNetwork
Contributing to ProgLearn
Submitting a bug report or a feature request
How to make a good bug report
Contributing Code
Pull Request Checklist
Coding Guidelines
Docstring Guidelines
Tutorial Guidelines
License
Useful Links
proglearn @ GitHub
proglearn @ PyPi
Issue Tracker
ProgLearn
Docs
»
Index
Index
A
|
C
|
F
|
G
|
K
|
L
|
N
|
P
|
S
|
T
|
U
A
add_task() (proglearn.ClassificationProgressiveLearner method)
(proglearn.LifelongClassificationForest method)
(proglearn.LifelongClassificationNetwork method)
add_transformer() (proglearn.ClassificationProgressiveLearner method)
(proglearn.LifelongClassificationForest method)
(proglearn.LifelongClassificationNetwork method)
C
ClassificationProgressiveLearner (class in proglearn)
F
fit() (proglearn.KNNClassificationVoter method)
(proglearn.NeuralClassificationTransformer method)
(proglearn.SimpleArgmaxAverage method)
(proglearn.TreeClassificationTransformer method)
(proglearn.TreeClassificationVoter method)
fit_transform() (proglearn.NeuralClassificationTransformer method)
(proglearn.TreeClassificationTransformer method)
G
generate_gaussian_parity() (in module proglearn.sims)
generate_spirals() (in module proglearn.sims)
get_params() (proglearn.KNNClassificationVoter method)
(proglearn.NeuralClassificationTransformer method)
(proglearn.SimpleArgmaxAverage method)
(proglearn.TreeClassificationTransformer method)
(proglearn.TreeClassificationVoter method)
get_task_ids() (proglearn.ClassificationProgressiveLearner method)
(proglearn.LifelongClassificationForest method)
(proglearn.LifelongClassificationNetwork method)
get_transformer_ids() (proglearn.ClassificationProgressiveLearner method)
(proglearn.LifelongClassificationForest method)
(proglearn.LifelongClassificationNetwork method)
K
KNNClassificationVoter (class in proglearn)
L
LifelongClassificationForest (class in proglearn)
(class in proglearn.forest)
LifelongClassificationNetwork (class in proglearn)
(class in proglearn.network)
N
NeuralClassificationTransformer (class in proglearn)
P
predict() (proglearn.ClassificationProgressiveLearner method)
(proglearn.KNNClassificationVoter method)
(proglearn.LifelongClassificationForest method)
(proglearn.LifelongClassificationNetwork method)
(proglearn.SimpleArgmaxAverage method)
(proglearn.TreeClassificationVoter method)
predict_proba() (proglearn.ClassificationProgressiveLearner method)
(proglearn.KNNClassificationVoter method)
(proglearn.LifelongClassificationForest method)
(proglearn.LifelongClassificationNetwork method)
(proglearn.SimpleArgmaxAverage method)
(proglearn.TreeClassificationVoter method)
S
score() (proglearn.KNNClassificationVoter method)
(proglearn.SimpleArgmaxAverage method)
(proglearn.TreeClassificationVoter method)
set_decider() (proglearn.ClassificationProgressiveLearner method)
(proglearn.LifelongClassificationForest method)
(proglearn.LifelongClassificationNetwork method)
set_params() (proglearn.KNNClassificationVoter method)
(proglearn.NeuralClassificationTransformer method)
(proglearn.SimpleArgmaxAverage method)
(proglearn.TreeClassificationTransformer method)
(proglearn.TreeClassificationVoter method)
set_transformer() (proglearn.ClassificationProgressiveLearner method)
(proglearn.LifelongClassificationForest method)
(proglearn.LifelongClassificationNetwork method)
set_voter() (proglearn.ClassificationProgressiveLearner method)
(proglearn.LifelongClassificationForest method)
(proglearn.LifelongClassificationNetwork method)
SimpleArgmaxAverage (class in proglearn)
T
transform() (proglearn.NeuralClassificationTransformer method)
(proglearn.TreeClassificationTransformer method)
TreeClassificationTransformer (class in proglearn)
TreeClassificationVoter (class in proglearn)
U
UncertaintyForest (class in proglearn.forest)