def test_KernelSmoothedFromBatchClassificationResult(self):
"""Kernel Smoothed Probability density graph from a single split"""
testfilename = 'test_graph_kernel_smoothed.npy'
graph = PredictedValuesGraph(self.batch_result)
graph.KernelSmoothedDensityGraph()
self.CompareGraphs(graph, testfilename)
def test_ErrMsgIfMatplotibNotInstalled(self):
"""Fail gracefully with informative message if matplotlib"""
graph = PredictedValuesGraph(self.batch_result)
with self.assertRaises(ImportError):
graph.RankOrderedPredictedValuesGraph()
with self.assertRaises(ImportError):
graph.KernelSmoothedDensityGraph()
rank_ordered_graphs = []
ks_density_graphs = []
norm_full_set = full_set.Normalize()
# arg _all=True performs rank order sort without reducing number of features
sorted_full_set_pearson_weights = PearsonFeatureWeights.NewFromFeatureSpace( norm_full_set ).Threshold( _all=True )
for bin_index in range( num_bins ):
name = "{0}\nFeatures ranked {1}-{2}".format( full_set.source_path,
bin_offset + 1, bin_offset + num_features_per_bin )
print name
sliced_pearson_weights = sorted_full_set_pearson_weights.Slice( bin_offset, bin_offset + num_features_per_bin )
sliced_fs = full_set.FeatureReduce( sliced_pearson_weights )
experiment = FeatureSpaceRegressionExperiment.NewShuffleSplit( sliced_fs )
grapher = PredictedValuesGraph( experiment )
grapher.RankOrderedPredictedValuesGraph( name )
grapher.SaveToFile( "rank_ordered_features_{0:03d}-{1:03d}".format( bin_offset + 1, bin_offset + num_features_per_bin ) )
grapher.KernelSmoothedDensityGraph( name )
grapher.SaveToFile( "ks_density_features_{0:03d}-{1:03d}".format( bin_offset + 1, bin_offset + num_features_per_bin ) )
bin_offset += num_features_per_bin
import subprocess
subprocess.call( [ "convert", "rank_ordered_features*", "rank_ordered.pdf"] )
subprocess.call( [ "convert", "ks_density_features*", "ks_density.pdf"] )