error in RF
Closed this issue · 1 comments
xuefzhao commented
we ran into this error during RF:
Labeling training data...
Traceback (most recent call last):
File "/apps/lab/miket/anaconda/4.3.30/envs/pysam_XZ/bin/svtk", line 6, in <module>
exec(compile(open(__file__).read(), __file__, 'exec'))
File "/data/talkowski/xuefang/local/src/svtk/scripts/svtk", line 67, in <module>
main()
File "/data/talkowski/xuefang/local/src/svtk/scripts/svtk", line 64, in main
getattr(cli, command)(sys.argv[2:])
File "/data/talkowski/xuefang/local/src/svtk/svtk/cli/adjudicate.py", line 35, in main
scores, cutoffs = adjudicate_SV(metrics)
File "/data/talkowski/xuefang/local/src/svtk/svtk/adjudicate/adjudicate_sv.py", line 331, in adjudicate_SV
cutoffs[6] = adjudicate_PESR(metrics)
File "/data/talkowski/xuefang/local/src/svtk/svtk/adjudicate/adjudicate_sv.py", line 249, in adjudicate_PESR
labeler, cutoffs, 'PESR_prob')
File "/data/talkowski/xuefang/local/src/svtk/svtk/adjudicate/random_forest.py", line 24, in rf_classify
rf.run()
File "/data/talkowski/xuefang/local/src/svtk/svtk/adjudicate/random_forest.py", line 63, in run
self.label_training_data()
File "/data/talkowski/xuefang/local/src/svtk/svtk/adjudicate/random_forest.py", line 74, in label_training_data
self.clean['label'] = self.labeler.label(self.clean)
File "/data/talkowski/xuefang/local/src/svtk/svtk/adjudicate/labelers.py", line 19, in label
return metrics.apply(self.label_row, axis=1)
File "/apps/lab/miket/anaconda/4.3.30/envs/pysam_XZ/lib/python3.6/site-packages/pandas/core/frame.py", line 4877, in apply
ignore_failures=ignore_failures)
File "/apps/lab/miket/anaconda/4.3.30/envs/pysam_XZ/lib/python3.6/site-packages/pandas/core/frame.py", line 4973, in _apply_standard
results[i] = func(v)
File "/data/talkowski/xuefang/local/src/svtk/svtk/adjudicate/labelers.py", line 108, in label_row
row.SR2_prob < 0.1):
File "/apps/lab/miket/anaconda/4.3.30/envs/pysam_XZ/lib/python3.6/site-packages/pandas/core/generic.py", line 3614, in __getattr__
return object.__getattribute__(self, name)
AttributeError: ("'Series' object has no attribute 'SR2_prob'", 'occurred at index 2')
quick fix we did: we modified SR2 to SR1