genome-windows-generator

![](https://github.com/zommiommy/genome_windows_generator/workflows/Python%20package/badge.svg)


License
MIT
Install
pip install genome-windows-generator==1.1.0

Documentation

genome_windows_generator

Travis CI build SonarCloud Quality SonarCloud Maintainability Codacy Maintainability Maintainability Pypi project Pypi total project downloads

How do I install this package?

As usual, just download it using pip:

pip install genome_windows_generator

Tests Coverage

Since some software handling coverages sometime get slightly different results, here's three of them:

Coveralls Coverage SonarCloud Coverage Code Climate Coverate

from genome_windows_generator import GenomeWindowsGenerator, NoisyWindowsGenerator

dg = NoisyWindowsGenerator(
    assembly="hg19",
    window_size=200,
    batch_size=3,
    buffer_size=5,
    test_chromosomes=["chr1", "chr5"]
)

The methods train, test returns two independant generator of the train data and test data respectivly.

This is package is mainly meant to be used with keras's fit_generator.

model.fit_generator(
    epochs=100,
    generator=dg.generator(),
    steps_per_epoch=dg.steps_per_epoch(),
    validation_data=dg.validation_data(),
    validation_steps=dg.validation_steps(),
)