Carlos-Francisco Méndez-Cruz / nlp-preprocessing-pipeline

Go to a project
Toggle navigation Toggle navigation pinning
Carlos-Francisco Méndez-Cruz
f972fd01 1 parent e425a3be

New terminological tagging for CRFs

Inline Side-by-side
Showing 1 changed file with 3 additions and 1 deletions
...@@ -85,9 +85,11 @@ if __name__ == "__main__": ...@@ -85,9 +85,11 @@ if __name__ == "__main__":
85 85
86 hashTermFiles = lists["hashTermFiles"] 86 hashTermFiles = lists["hashTermFiles"]
87 hashTerms = lists["hashTerms"] 87 hashTerms = lists["hashTerms"]
88 - hashTermsOrig = lists["hashTerms"] 88 + hashTermsOrig = {}
89 89
90 for key in hashTermFiles.keys(): 90 for key in hashTermFiles.keys():
91 + if key not in hashTermsOrig:
92 + hashTermsOrig[key] = []
91 for f in hashTermFiles[key]: 93 for f in hashTermFiles[key]:
92 # print('File: ' + f) 94 # print('File: ' + f)
93 with open(os.path.join(options.termPath, f), "r", encoding="utf-8", errors="replace") as iFile: 95 with open(os.path.join(options.termPath, f), "r", encoding="utf-8", errors="replace") as iFile:
......