Example running pipeline from script

This notebook gives an example of how the package can be used for running it interactively using a notebook.

First, import the package

import dvb.datascience as ds

C:\ProgramData\Anaconda3\lib\site-packages\deap\tools\_hypervolume\pyhv.py:33: ImportWarning: Falling back to the python version of hypervolume module. Expect this to be very slow.
  "module. Expect this to be very slow.", ImportWarning)
C:\ProgramData\Anaconda3\lib\importlib\_bootstrap_external.py:426: ImportWarning: Not importing directory C:\ProgramData\Anaconda3\lib\site-packages\mpl_toolkits: missing __init__
  _warnings.warn(msg.format(portions[0]), ImportWarning)
C:\ProgramData\Anaconda3\lib\importlib\_bootstrap_external.py:426: ImportWarning: Not importing directory c:\programdata\anaconda3\lib\site-packages\mpl_toolkits: missing __init__
  _warnings.warn(msg.format(portions[0]), ImportWarning)

Next, use the package to run the file ‘example.py’ (from the same directory as this notebook) using the run() method of the example.py file

p = ds.run_module('example').run()

Experiment started

Running pipeline 1

'Drawing diagram using blockdiag'

Running pipeline 2

'Drawing diagram using blockdiag'

Transform fit

'auc() not yet implemented for multiclass classifiers'

'plot_auc() not yet implemented for multiclass classifiers'

accuracy

fit
0.646667

mcc

fit
0.57563

Confusion Matrix

'Precision-recall-curve not yet implemented for multiclass classifiers'

'log_loss() not yet implemented for multiclass classifiers'

Classification Report

	precision	recall	f1-score	support
Setosa	0.50	1.00	0.666667	50.0
Versicolour	0.94	0.94	0.940000	50.0
Virginica	0.00	0.00	0.000000	50.0
avg/total	0.48	0.65	0.540000	150.0

Model Performance

Experiment done