Model Tooling library

Installation

Use pip to install: pip install ml-tooling Or use conda conda install -c conda-forge ml_tooling

Test

We use tox for managing build and test environments, to install tox run: pip install tox And to run tests: tox -e py

Example usage

Define a class using ModelData and implement the two required methods. Here we simply implement a linear regression on the Boston dataset using sklearn.datasets

from sklearn.datasets import fetch_california_housing
from sklearn.linear_model import LinearRegression

from ml_tooling import Model
from ml_tooling.data import Dataset

# Define a new data class
class CaliforniaData(Dataset):
    def load_prediction_data(self, idx):
        x, _ = fetch_california_housing(return_X_y=True)
        return x[idx] # Return given observation

    def load_training_data(self):
        return fetch_california_housing(return_X_y=True)

# Instantiate a model with an estimator
linear_california = Model(LinearRegression())

# Instantiate the data
data = CaliforniaData()

# Split training and test data
data.create_train_test()

# Score the estimator yielding a Result object
result = linear_california.score_estimator(data)

# Visualize the result
result.plot.prediction_error()

print(result)
<Result LinearRegression: {'r2': 0.68}>

Links

• Documentation: https://ml-tooling.readthedocs.io
• Releases: https://pypi.org/project/ml_tooling/
• Code: https://github.com/andersbogsnes/ml_tooling
• Issue Tracker: https://github.com/andersbogsnes/ml_tooling/issues