Skip to content

Releases: EpistasisLab/Aliro

v0.21.1

04 Dec 23:21
@jay-m-dev jay-m-dev
v0.21.1
dd99298
Compare
Choose a tag to compare
v0.21.1 Latest
Latest
  • Delete dataset icon was updated to a semantic-ui icon
  • The AliroEd intropage was updated
  • Bug fixes

What's Changed

Full Changelog: v1.0.1...v0.21.1

Contributors

jay-m-dev, dependabot, and HyunjunA
Assets 3
Loading

v0.21

11 Oct 00:48
@jay-m-dev jay-m-dev
v0.21
35ce9de
Compare
Choose a tag to compare
v0.21

Highlights in this release

  • Aliro v0.21 now allows deleting individual datasets.

See the documentation at https://epistasislab.github.io/Aliro/ for more instructions.

Requirements:

See the installation instructions for prerequisite software requirements.

Installation:

  1. Download the production zip Aliro-0_21.zip from the Assets section below (be sure not to download the source code zip).
  2. Unzip the archive
  3. Follow the steps in the Running section below.

Running:

See Using Aliro for instructions.

  • From the command line, navigate to the Aliro directory and run the command docker-compose up to start the Aliro server.
  • To stop Aliro, kill the process with ctrl+c and wait for the server to shut down.
  • Once the webserver is up, connect to http://localhost:5080/ to access the website.

What's Changed

Full Changelog: v0.20...v0.21

Contributors

jay-m-dev, dependabot, and HyunjunA
Assets 3
Loading

v0.20

19 Sep 01:48
@jay-m-dev jay-m-dev
v0.20
803b64a
Compare
Choose a tag to compare
v0.20

Changelog

  • Aliro v0.20 adds the ability to chat with the OpenAI API. Talk to a Large Language Model (LLM) about your dataset and discuss the experiment results. Plus, you can also request that the LLM provide custom scripts to modify or clean up your dataset.
  • Docker images are now built using docker buildx, which is used to build multi-architecture (amd64, arm64) images with a single command.
    • These multi-architecture versions are uploaded to Docker Hub. This means that this same build can be run on amd64 processors and arm64 (e.g. M1 Mac and Raspberry Pi)

See the documentation at https://epistasislab.github.io/Aliro/ for more instructions.

Requirements:

See the installation instructions for prerequisite software requirements.

Installation:

  1. Download the production zip Aliro-0_20.zip from the Assets section below (be sure not to download the source code zip).
  2. Unzip the archive
  3. Follow the steps in the Running section below.

Running:

See Using Aliro for instructions.

  • From the command line, navigate to the Aliro directory and run the command docker-compose up to start the Aliro server.
  • To stop Aliro, kill the process with ctrl+c and wait for the server to shut down.
  • Once the webserver is up, connect to http://localhost:5080/ to access the website.

What's Changed

Read more

Contributors

nickotto, jay-m-dev, and 2 other contributors
Assets 3
Loading

Aliro v0.19

10 Dec 01:57
@jay-m-dev jay-m-dev
v0.19
d109c72
Compare
Choose a tag to compare
Aliro v0.19

Changelog

  • Data visualization improvements including interactive and responsive t-SNE, PCA, feature importance, and learning curve plots for classification tasks
  • Add popups to show step-by-step instructions for using Aliro
  • Create Aliro Ed intro webpage and production webpage
  • Various bugfixes

See the documentation at https://epistasislab.github.io/Aliro/ for more instructions.

Requirements:

See the installation instructions for prerequisite software requirements.

Installation:

  1. Download the production zip Aliro-0_19.zip from the Assets section below (be sure not to download the source code zip).
  2. Unzip the archive

Running:

See Using Aliro for instructions.

  • From the command line, navigate to the Aliro directory and run the command docker-compose up to start the Aliro server.
  • To stop Aliro, kill the process with ctrl+c and wait for the server to shut down.
  • Once the webserver is up, connect to http://localhost:5080/ to access the website.

What's Changed

  • Add support for running PennAI on Raspberry Pi by @JDRomano2 in #305

  • Create aliroed web application with data visualization in D3.js by @HyunjunA in #376

  • Add popups to explain how to use aliro and interactive donut chart to show class rate information by @HyunjunA in #380

  • Add interactive t-sne, pca, feature important, and learning curve charts to aliro by @HyunjunA in #395

  • Update skl_util.py code to deal with exceptional case for learning curve plot by @HyunjunA in #438

  • Infvisfrontendmlbackend by @jay-m-dev in #444

New Contributors

Full Changelog: v0.17...v0.19

Contributors

JDRomano2, jay-m-dev, and HyunjunA
Assets 4
Loading

Aliro v0.18

23 Aug 18:49
@jay-m-dev jay-m-dev
v0.18
34c5ddc
Compare
Choose a tag to compare
Aliro v0.18

Changelog

  • Rebranding: PennAI is now called Aliro.
  • Updated NodeJS (version 12.22.9)

See the documentation at https://epistasislab.github.io/Aliro/ for more instructions.

Requirements:

See the installation instructions for prerequisite software requirements.

Installation:

  1. Download the production zip Aliro-0_18.zip from the Assets section below (be sure not to download the source code zip).
  2. Unzip the archive

Running:

See Using Aliro for instructions.

  • From the command line, navigate to the Aliro directory and run the command docker-compose up to start the Aliro server.
  • To stop Aliro, kill the process with ctrl+c and wait for the server to shut down.
  • Once the webserver is up, connect to http://localhost:5080/ to access the website.

What's Changed

Full Changelog: v0.17...v0.18

Contributors

mgstauffer, hjwilli, and 4 other contributors
Assets 3
Loading

Aliro for the Raspberry Pi (ARM64)

23 Aug 23:44
@jay-m-dev jay-m-dev
vrpi-0.18
f82a147
Compare
Choose a tag to compare
Aliro for the Raspberry Pi (ARM64)

This is the Raspberry Pi (ARM64) version of Aliro.

What's Changed

Full Changelog: v0.17...vrpi-0.18

Contributors

mgstauffer, JDRomano2, and 5 other contributors
Assets 3
Loading

v0.17

03 Dec 00:06
@hjwilli hjwilli
v0.17
858f91d
Compare
Choose a tag to compare
v0.17

Changelog

  • Released the PennAiPy python package - A scikit-learn API interface that allows PennAI to be used like a scikit learn estimator. Documentation can be found here: https://epistasislab.github.io/pennai/userguide_sklearn_api.html
  • Trained PennAI recommenders can be saved/loaded
  • Added default SVD PennAI recommenders trained on the PMLB benchmark suite
  • UI improvements, including improved dataset upload interface
  • Added SHAP analysis plots for classification experiments
  • Various bugfixes

See the documentation at https://epistasislab.github.io/pennai/ for more instructions.

Requirements:

See the installation instructions for prerequisite software requirements.

Installation:

  1. Download the production zip pennai-0_17.zip from the Assets section below (be sure not to download the source code zip).
  2. Unzip the archive

Running:

See Using PennAI for instructions.

  • From the command line, navigate to the pennai directory and run the command docker-compose up to start the PennAI server.
  • To stop PennAI, kill the process with ctrl+c and wait for the server to shut down.
  • Once the webserver is up, connect to http://localhost:5080/ to access the website.
Assets 3
Loading

Feature Updates

09 Apr 23:46
@hjwilli hjwilli
v0.16
32c772c
Compare
Choose a tag to compare
Feature Updates Pre-release
Pre-release

Changelog

  • AI toggle is disabled while the recommender engine is initializing
  • Classification knowledgebases updated
  • Prevent invalid ML combinations from being recommended
  • Invalidate browser cache when version changes
  • npm and python package updates

See the documentation at https://epistasislab.github.io/pennai/ for more instructions.

Requirements:

See the installation instructions for prerequisite software requirements.

Installation:

  1. Download the production zip pennai-0_15.zip from the Assets section below (be sure not to download the source code zip).
  2. Unzip the archive

Running:

See Using PennAI for instructions.

  • From the command line, navigate to the pennai directory and run the command docker-compose up to start the PennAI server.
  • To stop PennAI, kill the process with ctrl+c and wait for the server to shut down.
  • Once the webserver is up, connect to http://localhost:5080/ to access the website.
Assets 3
Loading

Feature Updates

06 Mar 21:27
@hjwilli hjwilli
v0.15
5809742
Compare
Choose a tag to compare
Feature Updates Pre-release
Pre-release

Changelog

  • Added support for regression problems
  • Maximum allowed file size can be set as a configuration option
  • Improved web caching support

See the documentation at https://epistasislab.github.io/pennai/ for more instructions.

Requirements:

See the installation instructions for prerequisite software requirements.

Installation:

  1. Download the production zip pennai-0_15.zip from the Assets section below (be sure not to download the source code zip).
  2. Unzip the archive

Running:

See Using PennAI for instructions.

  • From the command line, navigate to the pennai directory and run the command docker-compose up to start the PennAI server.
  • To stop PennAI, kill the process with ctrl+c and wait for the server to shut down.
  • Once the webserver is up, connect to http://localhost:5080/ to access the website.
Assets 3
Loading

Feature updates

18 Feb 22:35
@hjwilli hjwilli
v0.14
8974b63
Compare
Choose a tag to compare
Feature updates Pre-release
Pre-release

Changelog

  • Additional AI recommenders
  • Updated figures and metrics for classification experiments
  • Additional AI termination conditions
  • Support for multiple knowledgebase files
  • Improved scaling support for multiple 'machine' container instances
  • Refactored metafeature/internal id tracking
  • Numerous bugfixes and UI improvements

See the documentation at https://epistasislab.github.io/pennai/ for more instructions.

Requirements:

See the installation instructions for prerequisite software requirements.

Installation:

  1. Download the production zip pennai-0_14.zip from the Assets section below (be sure not to download the source code zip).
  2. Unzip the archive

Running:

See Using PennAI for instructions.

  • From the command line, navigate to the pennai directory and run the command docker-compose up to start the PennAI server.
  • To stop PennAI, kill the process with ctrl+c and wait for the server to shut down.
  • Once the webserver is up, connect to http://localhost:5080/ to access the website.
Assets 3
Loading