Skip to main content

Just another ML framework. Built on top of scikit-learn.

Project description



General use, pipeline-oriented ML platform built on top of sklearn. Lets user configure pipelines, load data, and evaluate pipeline against data. Who's said that improvising over Giant Steps has to be hard? This framework eases and standardizes research process. User can focus on configuring pipeline, or implementing core pipeline elements.


  • processing research batches
  • some nice plots by plotly
  • data wrappers and loaders

Data set loaders

Plain .csv

  • single label

Getting started

pip install coltrane


Load data set.

from pathlib import Path

from coltrane.classification import Inspector
from import Data

__LOG = Path('log')
__DATA_IRIS = Path('coltrane/test/data/iris.csv')
__RANDOM_STATE = 45625461

data = Data(path=__DATA_IRIS)

Create Inspector instance and run inspection. This will output some stats to provided output directory.

inspector = Inspector()
summary = inspector.inspect(data, output=__LOG)

You can access some additional stats and plot them.

# this one's useful for cost-sensitive learning
class_balance = inspector.get_class_balance(data)

Check notebook examples for more.


from pathlib import Path

from sklearn.metrics import (accuracy_score, f1_score, make_scorer,
                             precision_score, recall_score)
from sklearn.model_selection import RepeatedStratifiedKFold
from sklearn.naive_bayes import GaussianNB
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import LabelEncoder, StandardScaler

from coltrane import Batch
from coltrane.classification import Processor
from import Data

__LOG = Path('log')
__DATA_IRIS = Path('coltrane/test/data/iris.csv')
__RANDOM_STATE = 45625461

batch = Batch(
            ('standard-scaler', StandardScaler()),
            ('naive-bayes', GaussianNB())
        'accuracy': make_scorer(accuracy_score),
        'precision': make_scorer(precision_score, average='macro'),
        'recall': make_scorer(recall_score, average='macro'),
        'f1': make_scorer(f1_score, average='macro')

processor = Processor()
stats = processor.process(batch, output=__LOG)


Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for coltrane, version 0.1.2
Filename, size File type Python version Upload date Hashes
Filename, size coltrane-0.1.2-py3-none-any.whl (15.1 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size coltrane-0.1.2.tar.gz (10.0 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page