Tools for exploratory data analysis in Python
Python
Switch branches/tags
Nothing to show
Clone or download
Latest commit 4526d29 Mar 30, 2017
Permalink
Failed to load latest commit information.
Dora bump version for pip fix Feb 24, 2016
.gitignore .gitignore and setup.py Feb 24, 2016
README.md reformat readme Mar 30, 2017
setup.py bump version for pip fix Feb 24, 2016

README.md

Dora

Exploratory data analysis toolkit for Python.

Contents

Summary

Dora is a Python library designed to automate the painful parts of exploratory data analysis.

The library contains convenience functions for data cleaning, feature selection & extraction, visualization, partitioning data for model validation, and versioning transformations of data.

The library uses and is intended to be a helpful addition to common Python data analysis tools such as pandas, scikit-learn, and matplotlib.

Setup

$ pip3 install Dora
$ python3
>>> from Dora import Dora

Usage

Reading Data & Configuration

# without initial config
>>> dora = Dora()
>>> dora.configure(output = 'A', data = 'path/to/data.csv')

# is the same as
>>> import pandas as pd
>>> dataframe = pd.read_csv('path/to/data.csv')
>>> dora = Dora(output = 'A', data = dataframe)

>>> dora.data
   A   B  C      D  useless_feature
0  1   2  0   left                1
1  4 NaN  1  right                1
2  7   8  2   left                1

Cleaning

# read data with missing and poorly scaled values
>>> import pandas as pd
>>> df = pd.DataFrame([
...   [1, 2, 100],
...   [2, None, 200],
...   [1, 6, None]
... ])
>>> dora = Dora(output = 0, data = df)
>>> dora.data
   0   1    2
0  1   2  100
1  2 NaN  200
2  1   6  NaN

# impute the missing values (using the average of each column)
>>> dora.impute_missing_values()
>>> dora.data
   0  1    2
0  1  2  100
1  2  4  200
2  1  6  150

# scale the values of the input variables (center to mean and scale to unit variance)
>>> dora.scale_input_values()
>>> dora.data
   0         1         2
0  1 -1.224745 -1.224745
1  2  0.000000  1.224745
2  1  1.224745  0.000000

Feature Selection & Extraction

# feature selection / removing a feature
>>> dora.data
   A   B  C      D  useless_feature
0  1   2  0   left                1
1  4 NaN  1  right                1
2  7   8  2   left                1

>>> dora.remove_feature('useless_feature')
>>> dora.data
   A   B  C      D
0  1   2  0   left
1  4 NaN  1  right
2  7   8  2   left

# extract an ordinal feature through one-hot encoding
>>> dora.extract_ordinal_feature('D')
>>> dora.data
   A   B  C  D=left  D=right
0  1   2  0       1        0
1  4 NaN  1       0        1
2  7   8  2       1        0

# extract a transformation of another feature
>>> dora.extract_feature('C', 'twoC', lambda x: x * 2)
>>> dora.data
   A   B  C  D=left  D=right  twoC
0  1   2  0       1        0     0
1  4 NaN  1       0        1     2
2  7   8  2       1        0     4

Visualization

# plot a single feature against the output variable
dora.plot_feature('column-name')

# render plots of each feature against the output variable
dora.explore()

Model Validation

# create random partition of training / validation data (~ 80/20 split)
dora.set_training_and_validation()

# train a model on the data
X = dora.training_data[dora.input_columns()]
y = dora.training_data[dora.output]

some_model.fit(X, y)

# validate the model
X = dora.validation_data[dora.input_columns()]
y = dora.validation_data[dora.output]

some_model.score(X, y)

Data Versioning

# save a version of your data
>>> dora.data
   A   B  C      D  useless_feature
0  1   2  0   left                1
1  4 NaN  1  right                1
2  7   8  2   left                1
>>> dora.snapshot('initial_data')

# keep track of changes to data
>>> dora.remove_feature('useless_feature')
>>> dora.extract_ordinal_feature('D')
>>> dora.impute_missing_values()
>>> dora.scale_input_values()
>>> dora.data
   A         B         C    D=left   D=right
0  1 -1.224745 -1.224745  0.707107 -0.707107
1  4  0.000000  0.000000 -1.414214  1.414214
2  7  1.224745  1.224745  0.707107 -0.707107

>>> dora.logs
["self.remove_feature('useless_feature')", "self.extract_ordinal_feature('D')", 'self.impute_missing_values()', 'self.scale_input_values()']

# use a previous version of the data
>>> dora.snapshot('transform1')
>>> dora.use_snapshot('initial_data')
>>> dora.data
   A   B  C      D  useless_feature
0  1   2  0   left                1
1  4 NaN  1  right                1
2  7   8  2   left                1
>>> dora.logs
[]

# switch back to your transformation
>>> dora.use_snapshot('transform1')
>>> dora.data
   A         B         C    D=left   D=right
0  1 -1.224745 -1.224745  0.707107 -0.707107
1  4  0.000000  0.000000 -1.414214  1.414214
2  7  1.224745  1.224745  0.707107 -0.707107
>>> dora.logs
["self.remove_feature('useless_feature')", "self.extract_ordinal_feature('D')", 'self.impute_missing_values()', 'self.scale_input_values()']

Testing

To run the test suite, simply run python3 spec.py from the Dora directory.

Contribute

Pull requests welcome! Feature requests / bugs will be addressed through issues on this repository. While not every feature request will necessarily be handled by me, maintaining a record for interested contributors is useful.

Additionally, feel free to submit pull requests which add features or address bugs yourself.

License

The MIT License (MIT)

Copyright (c) 2016 Nathan Epstein

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.