Companion code for "Bayesian logistic regression for online recalibration and revision of risk prediction models with performance guarantees"

Overview

Companion code for "Bayesian logistic regression for online recalibration and revision of risk prediction models with performance guarantees"

Installation

We use pip to install things into a python virtual environment. Refer to requirements.txt for package requirements. We use nestly + SCons to run simulations.

File descriptions

generate_data_single_pop.py -- Simulate a data stream from a single population following a logistic regression model.

  • Inputs:
    • --simulation: string for selecting the type of distribution shift. Options for this argument are the keys in SIM_SETTINGS in constants.py.
  • Outputs:
    • --out-file: pickle file containing the data stream

generate_data_two_pop.py -- Simulate a data stream from two subpopulations, where each are generated using logistic regression models. Similar arguments as generate_data_single_pop.py. The percentage split beween the two subpopulations is controlled by the --subpopulations argument.

  • Outputs:
    • --out-file: pickle file containing the data stream

create_modeler.py -- Creates a model developer who fits the original prediction model and may propose a continually refitted model at each time point.

  • Inputs:
    • --data-file: pickle file with the entire data stream
    • --simulation: string for selecting the model refitting strategy by the model developer. Options are to keep the model locked (locked), refit on all accumulated data (cumulative_refit), and refit on the latest observations within some window length (boxed, window length specified by --max-box). The last two options is to train an ensemble with the original and the cumulative_refit models (combo_refit) and train an ensemble with the original and the boxed models (combo_boxed).
  • Outputs:
    • --out-file: pickle file containing the modeler

main.py -- Given the data and the model developer, run online model recalibration/revision using MarBLR and BLR.

  • Inputs:
    • --data-file: pickle file with the entire data stream
    • --model-file: pickle file with the model developer
    • --type-i-regret-factor: Type I regret will be controlled at the rate of args.type_i_regret_factor * (Initial loss of the original model)
    • --reference-recalibs: comma-separated string to select which other online model revisers to run. Options are no updating at all locked, ADAM adam, cumulative logistic regression cumulativeLR.
  • Outputs:
    • --obs-scores-file: csv file containing predicted probabilities and observed outcomes on the data stream
    • --history-file: csv file containing the predicted and actual probabilities on a held-out test data stream (only available if the data stream was simulated)
    • --scores-file: csv file containing performance measures on a held-out test data stream (only available if the data stream was simulated)
    • --recalibrators-file: pickle file containing the history of the online model revisers

Reproducing simulation results

The simulation_recalib folder contains the first set of simulations for online model recalibration. The simulation_revise folder contains the second set of simulations where we perform online logistic revision. The simulation_revise folder contains the third set of simulations where we perform online ensembling of the original model with a continually refitted model. The copd_analysis folder contains code for online model recalibration and revision for the COPD dataset. To reproduce the simulations, run scons .

Yoga - Yoga asana classifier for python

Yoga Asana Classifier Description Hi welcome to my new deep learning project "Yo

Programminghut 35 Dec 12, 2022
EZ graph is an easy to use AI solution that allows you to make and train your neural networks without a single line of code.

EZ-Graph EZ Graph is a GUI that allows users to make and train neural networks without writing a single line of code. Requirements python 3 pandas num

1 Jul 03, 2022
This repository contains several image-to-image translation models, whcih were tested for RGB to NIR image generation. The models are Pix2Pix, Pix2PixHD, CycleGAN and PointWise.

RGB2NIR_Experimental This repository contains several image-to-image translation models, whcih were tested for RGB to NIR image generation. The models

5 Jan 04, 2023
Creating predictive checklists from data using integer programming.

Learning Optimal Predictive Checklists A Python package to learn simple predictive checklists from data subject to customizable constraints. For more

Healthy ML 5 Apr 19, 2022
Auxiliary Raw Net (ARawNet) is a ASVSpoof detection model taking both raw waveform and handcrafted features as inputs, to balance the trade-off between performance and model complexity.

Overview This repository is an implementation of the Auxiliary Raw Net (ARawNet), which is ASVSpoof detection system taking both raw waveform and hand

6 Jul 08, 2022
Collapse by Conditioning: Training Class-conditional GANs with Limited Data

Collapse by Conditioning: Training Class-conditional GANs with Limited Data Moha

Mohamad Shahbazi 33 Dec 06, 2022
Galactic and gravitational dynamics in Python

Gala is a Python package for Galactic and gravitational dynamics. Documentation The documentation for Gala is hosted on Read the docs. Installation an

Adrian Price-Whelan 101 Dec 22, 2022
Python framework for Stochastic Differential Equations modeling

SDElearn: a Python package for SDE modeling This package implements functionalities for working with Stochastic Differential Equations models (SDEs fo

4 May 10, 2022
Dados coletados e programas desenvolvidos no processo de iniciação científica

Iniciacao_cientifica_FAPESP_2020-14845-6 Dados coletados e programas desenvolvidos no processo de iniciação científica Os arquivos .py são os programa

1 Jan 10, 2022
Tensorflow 2.x implementation of Panoramic BlitzNet for object detection and semantic segmentation on indoor panoramic images.

Deep neural network for object detection and semantic segmentation on indoor panoramic images. The implementation is based on the papers:

Alejandro de Nova Guerrero 9 Nov 24, 2022
[ICCV 2021] Self-supervised Monocular Depth Estimation for All Day Images using Domain Separation

ADDS-DepthNet This is the official implementation of the paper Self-supervised Monocular Depth Estimation for All Day Images using Domain Separation I

LIU_LINA 52 Nov 24, 2022
A Python framework for conversational search

Chatty Goose Multi-stage Conversational Passage Retrieval: An Approach to Fusing Term Importance Estimation and Neural Query Rewriting Installation Ma

Castorini 36 Oct 23, 2022
i-RevNet Pytorch Code

i-RevNet: Deep Invertible Networks Pytorch implementation of i-RevNets. i-RevNets define a family of fully invertible deep networks, built from a succ

Jörn Jacobsen 378 Dec 06, 2022
Code release for "Self-Tuning for Data-Efficient Deep Learning" (ICML 2021)

Self-Tuning for Data-Efficient Deep Learning This repository contains the implementation code for paper: Self-Tuning for Data-Efficient Deep Learning

THUML @ Tsinghua University 101 Dec 11, 2022
[TIP 2020] Multi-Temporal Scene Classification and Scene Change Detection with Correlation based Fusion

Multi-Temporal Scene Classification and Scene Change Detection with Correlation based Fusion Code for Multi-Temporal Scene Classification and Scene Ch

Lixiang Ru 33 Dec 12, 2022
Repositório para arquivos sobre o Módulo 1 do curso Top Coders da Let's Code + Safra

850-Safra-DS-ModuloI Repositório para arquivos sobre o Módulo 1 do curso Top Coders da Let's Code + Safra Para aprender mais Git https://learngitbranc

Brian Nunes 7 Dec 10, 2022
Monocular Depth Estimation Using Laplacian Pyramid-Based Depth Residuals

LapDepth-release This repository is a Pytorch implementation of the paper "Monocular Depth Estimation Using Laplacian Pyramid-Based Depth Residuals" M

Minsoo Song 205 Dec 30, 2022
Pytorch implementation of U-Net, R2U-Net, Attention U-Net, and Attention R2U-Net.

pytorch Implementation of U-Net, R2U-Net, Attention U-Net, Attention R2U-Net U-Net: Convolutional Networks for Biomedical Image Segmentation https://a

leejunhyun 2k Jan 02, 2023
Pretrained models for Jax/Haiku; MobileNet, ResNet, VGG, Xception.

Pre-trained image classification models for Jax/Haiku Jax/Haiku Applications are deep learning models that are made available alongside pre-trained we

Alper Baris CELIK 14 Dec 20, 2022
Aws-machine-learning-university-accelerated-tab - Machine Learning University: Accelerated Tabular Data Class

Machine Learning University: Accelerated Tabular Data Class This repository contains slides, notebooks, and datasets for the Machine Learning Universi

AWS Samples 916 Dec 23, 2022