I am stuck on an error for a new preprocessing submission to the Training queue. It appears the 'cancer' field (the data label) is missing, causing an error in my code. Note on line 2 of STDOUT below, the absence of the 'cancer' field from the first row of the images_crosswalk.tsv file:
STDOUT: 2017-03-08T21:44:53.958868599Z Treating : 000000.dcm
STDOUT: 2017-03-08T21:44:53.958894768Z {'laterality': 'R', 'subjectId': '1', 'filename': '000000.dcm', 'examIndex': '1', 'imageIndex': '1', 'view': 'CC'}
STDERR: 2017-03-08T21:44:53.958944509Z Traceback (most recent call last):
STDERR: 2017-03-08T21:44:53.958955321Z File "main.py", line 2, in
STDERR: 2017-03-08T21:44:53.958958611Z augment_data.augment_data()
STDERR: 2017-03-08T21:44:53.958960923Z File "/augment_data.py", line 36, in augment_data
STDERR: 2017-03-08T21:44:53.958980690Z if row['cancer'] == '1':
STDERR: 2017-03-08T21:44:53.958983172Z KeyError: 'cancer'
STDERR: 2017-03-08T21:44:54.014744229Z
STDERR: 2017-03-08T21:44:54.014770070Z real 0m2.846s
STDERR: 2017-03-08T21:44:54.014773368Z user 0m1.505s
STDERR: 2017-03-08T21:44:54.014775873Z sys 0m3.340s
This code works correctly on my machine. Any help would be greatly appreciated!
Created by jcb Hi jcb,
Please refer to the [Challenge Dictionary](https://www.synapse.org/#!Synapse:syn7214004) that describes the content of the exams metadata and images crosswalk files.
Thanks!