Add on for Hyperactive package to visualize progress of optimization run.
Project description
Progress Board
An addon for the Hyperactive package to visualize the progress of optimization runs.
The Progress Board is a dashboard (opens in webbrowser) that provides visualization of live-updated data from Hyperactive. It integrates seamlessly with Hyperactive (v4) and opens up the optimization run with useful information. It also supports multiprocessing and multiple searches at the same time without any added complexity or work for the user.
The Progress Board should be used for computationally expensive objective functions (like machine-/deep-learning models).
The Progress Board is tested in Ubuntu, but Windows support maybe added in the future.
Installation
pip install hyperactive-progress-board
Example
from sklearn.model_selection import cross_val_score
from sklearn.tree import DecisionTreeRegressor
from sklearn.datasets import fetch_california_housing
from hyperactive import Hyperactive
from hyperactive_progress_board import ProgressBoard # import progress board
data = fetch_california_housing()
X, y = data.data, data.target
progress = ProgressBoard() # init progress board
@progress.update # add decorator
def dtr_model(opt):
dtr = DecisionTreeRegressor(
min_samples_split=opt["min_samples_split"],
)
scores = cross_val_score(dtr, X, y, cv=5)
return scores.mean()
search_space = {
"max_depth": list(range(2, 50)),
"min_samples_split": list(range(2, 50)),
"min_samples_leaf": list(range(1, 50)),
}
progress.open() # open progress board before run begins
hyper = Hyperactive()
hyper.add_search(dtr_model, search_space, n_iter=1000)
hyper.run()
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
File details
Details for the file hyperactive_progress_board-0.1.0-py3-none-any.whl
.
File metadata
- Download URL: hyperactive_progress_board-0.1.0-py3-none-any.whl
- Upload date:
- Size: 11.1 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.26.0 setuptools/59.4.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.8.5
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 5f38cf1207bc3498cbf96cd775821a2f1ca6972305724ff76f56efb903fc5b70 |
|
MD5 | f4f3d988408c62517a1b40e4da15517b |
|
BLAKE2b-256 | 7dad619a93bbbd5f9508705b863bef90274c8bb8c4fd2d4ba10c63e85a2f104c |