bigbrain 0.0.2

A toolbox for learning anything and everything from data.

bigbrain

bigbrain is a toolbox for learning anything and everything from data.

No machine learning knowledge needed!

With this toolbox, all your data science problems can be solved with a single command. No longer will you need to know anything about machine learning in order to use its power to solve all your problems.

• Learn the best models from your data without any background knowledge!
• Let the AI learn the best algorithm for your data, and then use what it learned on other data!

Tech

bigbrain wraps scikit-learn to try many different models on your data and find the best performing algorithm for you.

Installation

bigbrain can be installed using pypi, either from the website, or from the command line:

$ pip install bigbrain

Example Usage

With this package, you can easily create regression or classification models for supervised learning techniques.

>>> from bigbrain import AI
>>> from sklearn.datasets import iris, load_boston
>>> from sklearn.model_selection import train_test_split
>>> from sklearn.metrics import accuracy_score,
...     mean_squared_error
####### Classification #######
>>> X, y = iris(return_X_y=True)
>>> X_train, X_test, y_train, y_test = train_test_split(
...     X, y, test_size=0.3)
>>> machine = AI(model_type='classification')
>>> machine.learn(X_train, y_train)
	"My big brain has learned everything."
>>> predictions = machine.go(X_test)
>>> print(accuracy_score(y_test, predictions)
	"0.97"
####### Regression #######
>>> X, y = load_boston(return_X_y=True)
>>> X_train, X_test, y_train, y_test = train_test_split(
...     X, y, test_size=0.3)
>>> machine = AI(model_type='regression')
"My big brain has learned everything."
>>> predictions = machine.go(X_test)
>>> print(mean_squared_error(y_test, predictions)
	"14.03"