Update process.py

process.py

@@ -127,7 +127,6 @@
     parser.add_argument('-do_cleanup', dest='do_cleanup', type=int, help='Wether or not to cleanup NERs that are just stopwords or small number', default = 1)
     parser.add_argument('-do_marian_mt', dest='do_marian_mt', type=int, help='Wether or not to use marianMT for translation instead of M2M100', default = 0)
     parser.add_argument('-do_docs_trim_for_person', dest='do_docs_trim_for_person', type=int, help='Wether or not to filter out documents with no mentions of persons', default = 0)
-    parser.add_argument('-do_docs_filter', dest='do_docs_filter', type=int, help='Wether or not to filter out documents with high ratios of junk, or CSAM', default = 0)
     parser.add_argument('-do_kenlm', dest='do_kenlm', type=int, help='Wether or not to apply a KenLM model to decide if a name is a common person name', default = 1)
     parser.add_argument('-do_qg_rel', dest='do_qg_rel', type=int, help='Wether or not to infer a relationship between PII entities based an question generation (EXPERIMENTAL)', default = 0)
     parser.add_argument('-num_words_per_chunk', dest='num_words_per_chunk', type=int, help='number of words per chunk', default=70)
@@ -258,7 +257,6 @@
                     regex_weight=args.regex_weight,
                     backtrans_weight=args.backtrans_weight,
                     do_docs_trim_for_person=args.do_docs_trim_for_person,
-                    do_docs_filter=args.do_docs_filter,
                     do_qg_rel=args.do_qg_rel,
                     do_kenlm = args.do_kenlm,
                     cutoff=cutoff,
@@ -290,7 +288,6 @@
                     regex_weight=args.regex_weight,
                     backtrans_weight=args.backtrans_weight,
                     do_docs_trim_for_person=args.do_docs_trim_for_person,
-                    do_docs_filter=args.do_docs_filter,
                     do_qg_rel=args.do_qg_rel,
                     do_kenlm = args.do_kenlm,
                     cutoff=cutoff,