adjustments to functions to make them accessible to AI classifier code

partition.py

@@ -8,17 +8,11 @@
 import matplotlib.pyplot as plt
 import numpy
 import spacy
-from utils import clean_corpus, reconstruct_hyphenated_words
+from utils import clean_corpus, reconstruct_hyphenated_words, create_sent_label_dict
 
 # Functions
 
-# Creates dictionary of a set, associating sentence with label
-def create_sent_label_dict(sents, labels):
-    sents_dict = []
-    for row_id,row in enumerate(sents):
-        row = re.sub("\n", " ", row)
-        sents_dict.append({"text":row.strip(), "label":labels[row_id]})
-    return sents_dict
+
 
 # Writes json of partition set, each entry is the sentence associated with its labels
 def write_partition_file(partition_dict, name):

simpleclassifier.py

@@ -70,9 +70,9 @@ def simple_classifier(sents_ref_json):
 test_pred_dict = simple_classifier(test_sents_ref_json)
 
 # Output predictions in separate files
-write_predictions_file("Train", train_pred_dict)
-write_predictions_file("Dev", dev_pred_dict)
-write_predictions_file("Test", test_pred_dict)
+write_predictions_file(train_pred_dict, 'output/Simple Classifier/multilabelPredictions_Train.json')
+write_predictions_file(dev_pred_dict,'output/Simple Classifier/multilabelPredictions_Dev.json')
+write_predictions_file(test_pred_dict, 'output/Simple Classifier/multilabelPredictions_Test.json')
 
 # Formatting predictions to calculate results
 train_pred_array = [sent['label'] for sent in train_pred_dict]