GitHub - oaoni/HyperInteractive: Interactive exploration of hyperparameter tuning results with ipywidget and plotly in jupyter notebook. · GitHub
Skip to content

oaoni/HyperInteractive

Folders and files

Repository files navigation

HyperInteractive

Interactive ipywidget and plotly framework for exploring hyperparameter tuning results

Hyper Explore Demo

Requirements

plotly == 4.12.0
ipywidgets == 7.5.1

Getting started

Clone repo and cd into the project directory

$ git clone https://github.com/oaoni/HyperInteractive.git
$ cd HyperInteractive

Launch in a classic jupyter notebook

$ jupyter notebook

Usage

import pandas as pd
from interactivehyper import hyperExplore

data = pd.read_csv('./demo/modeltune.csv')

initial_axis = ['best_test_loss','best_test_corr']
initial_surface_axis = ['mu','alpha','best_test_corr']
legend_group = 'model'
hover_items = ['learning_rate','alpha','mu']

tab = hyperExplore(data,initial_axis,initial_surface_axis,legend_group,hover_items)
tab

About

Interactive exploration of hyperparameter tuning results with ipywidget and plotly in jupyter notebook.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

Contributors

Languages