معرفی شرکت ها

تبلیغات ما

مشتریان به طور فزاینده ای آنلاین هستند. تبلیغات می تواند به آنها کمک کند تا کسب و کار شما را پیدا کنند.

مشاهده بیشتر

تبلیغات ما

مشتریان به طور فزاینده ای آنلاین هستند. تبلیغات می تواند به آنها کمک کند تا کسب و کار شما را پیدا کنند.

مشاهده بیشتر

تبلیغات ما

مشتریان به طور فزاینده ای آنلاین هستند. تبلیغات می تواند به آنها کمک کند تا کسب و کار شما را پیدا کنند.

مشاهده بیشتر

تبلیغات ما

مشتریان به طور فزاینده ای آنلاین هستند. تبلیغات می تواند به آنها کمک کند تا کسب و کار شما را پیدا کنند.

مشاهده بیشتر

تبلیغات ما

مشتریان به طور فزاینده ای آنلاین هستند. تبلیغات می تواند به آنها کمک کند تا کسب و کار شما را پیدا کنند.

مشاهده بیشتر

توضیحات

Dashboard to explore the data and to create baseline Machine Learning model.

ویژگی	مقدار
سیستم عامل	-
نام فایل	data-dashboard-0.1.1
نام	data-dashboard
نسخه کتابخانه	0.1.1
نگهدارنده	[]
ایمیل نگهدارنده	[]
نویسنده	Maciej Dowgird
ایمیل نویسنده	dowgird.maciej@gmail.com
آدرس صفحه اصلی	https://github.com/maciek3000/data_dashboard
آدرس اینترنتی	https://pypi.org/project/data-dashboard/
مجوز	-

# data-dashboard ## Short description Creates a simple static HTML dashboard with provided ``X``, ``y`` data to help users see what's going on in their data, assists in making decisions regarding features and finds the best "baseline" Machine Learning Model. ![data_dashboard](./docs/source/screenshots/dashboard.gif) ## Longer Description ### Installation and Usage ``data_dashboard`` library allows you to build HTML Dashboard visualizing not only the data and relationships between features but also automatically search for the best 'baseline' sklearn compatible Model. You can install the package via pip: ```pip install data-dashboard``` To create a Dashboard you need the data: ``X``, ``y`` and the ``output_directory`` where the HTML files will be placed. You can use toy datasets from ``examples`` (e.g. ``iris`` dataset) included in the library as well: ```python from data_dashboard import Dashboard from data_dashboard.examples import iris output_directory = "your_path/dashboard_output" X, y, descriptions = iris() # descriptions is additional argument described further in docs dsh = Dashboard(X, y, output_directory, descriptions) dsh.create_dashboard() ``` Created HTML Dashboard will contain 3 subpages for you to look: * Overview with summary statistics of the data * Features where you can dig deeper into each feature in the data * Models showing search results and models performances ### Description ``data_dashboard`` aims to help you in those initial moments when you get your hands on new data and you ask yourself a question "Now what?". Instead of going through Jupyter Notebook tables and visualizations, ``data_dashboard`` gathers all that information in one place in a user-friendly interface. What is more, automated 'baseline' sklearn Model is created - you can then adjust not only parameters and algorithms, but also see how your changes affect the performance. There is also an educational twist to ``data_dashboard``, as explanations on correlations and transformations might alleviate pains that beginner Data Scientists might encounter (e.g. what do Transformers really do with my features?). Furthermore, provided visualizations might also assist in manual feature engineering. Last but not least, ``data_dashboard`` tries to put more emphasis on the design of both HTML and Visualization elements. If you are not the biggest fan of default matplotlib plots, then you can give ``data_dashboard`` a try - perhaps the styling will suit your taste! ### Working Example Example with Titanic dataset can be found here: https://example-data-dashboard.herokuapp.com/ ### Documentation Documentation can be found here: <https://data-dashboard.readthedocs.io/> ### Author MIT License Copyright (c) 2021 Maciej Dowgird For contact use: [dowgird.maciej@gmail.com](mailto:dowgird.maciej@gmail.com)

نیازمندی

مقدار	نام
>=1.2.3	pandas
>=1.19.5	numpy
>=1.6.1	scipy
>=4.9.3	beautifulsoup4
>=0.24.1	scikit-learn
>=0.11.1	seaborn
>=2.3.0	bokeh
>=2.11.3	Jinja2
>=1.3.3	xgboost
>=3.2.0	lightgbm

زبان مورد نیاز

مقدار	نام
>=3.7	Python

نحوه نصب

نصب پکیج whl data-dashboard-0.1.1:

pip install data-dashboard-0.1.1.whl

نصب پکیج tar.gz data-dashboard-0.1.1:

pip install data-dashboard-0.1.1.tar.gz