def build_features(data_file, feat_name, binning):
# read text data
category_texts = data_helper.get_reviews(os.path.join("./", data_file))
# build features
features_category_tuples, texts = get_features_category_tuples(
category_texts, feat_name, binning)
return features_category_tuples, texts
def build_features(data_file, feat_name, save_feats=None, binning=False):
# read text liwc
positive_texts, negative_texts = data_helper.get_reviews(
os.path.join(DATA_DIR, data_file))
category_texts = {"positive": positive_texts, "negative": negative_texts}
# build features
features_category_tuples, texts = features.get_features_category_tuples(
category_texts, feat_name, data_file)
# save features to file
if save_feats is not None:
write_features_category(features_category_tuples, save_feats)
return features_category_tuples, texts
def build_features(data_file, feat_name, save_feats=None, binning=False):
# read text data
raw_data = data_helper.read_file(data_file)
positive_texts, negative_texts = data_helper.get_reviews(raw_data)
category_texts = {"positive": positive_texts, "negative": negative_texts}
# build features
features_category_tuples, texts = get_features_category_tuples(
category_texts, feat_name)
# save features to file
if save_feats is not None:
write_features_category(features_category_tuples, save_feats)
return features_category_tuples, texts
def build_features(data_file, feat_name, binning, save_feats=None, test=False):
# read text data
positive_texts, negative_texts = data_helper.get_reviews(os.path.join(DATA_DIR, data_file))
if test:
category_texts = {"unknown": positive_texts, "unknown2": negative_texts}
else:
category_texts = {"positive": positive_texts, "negative": negative_texts}
# build features
features_category_tuples, texts = get_features_category_tuples(category_texts, feat_name, binning)
# save features to file
if save_feats is not None:
write_features_category(features_category_tuples, save_feats, test)
return features_category_tuples, texts
def build_features(data_file, feat_name):
# read text data
positive_texts, negative_texts = data_helper.get_reviews(
os.path.join(DATA_DIR, data_file))
category_texts = {"positive": positive_texts, "negative": negative_texts}
# build features
features_category_tuples, texts = get_features_category_tuples(
category_texts, feat_name)
#labelDict = {}
#for item in features_category_tuples:
# for tup in item[0]: #go through items in dict
# if tup == "BIGRAM_service_friendly" and item[0][tup] == 1:
# print(tup)
# labelDict[(tup, item[0][tup])] = item[1]
#print(labelDict)
return features_category_tuples, texts
def build_features(data_file, feat_name):
# read text data
positive_texts, negative_texts = data_helper.get_reviews(
os.path.join(DATA_DIR, data_file))
category_texts = {"positive": positive_texts, "negative": negative_texts}
# build features
features_category_tuples, texts = get_features_category_tuples(
category_texts, feat_name)
# save features to file
datamap = {}
# print("Feature: " + str(feat_name)+"_ Data: "+str(data_file))
datamap["dev_examples.tsv"] = "development"
datamap["train_examples.tsv"] = "training"
write_features_category(features_category_tuples, str(
feat_name)+"-"+datamap[str(data_file)])
return features_category_tuples, texts
def build_features(data_file, feat_name, save_feats=None):
# read text data
if data_file == "test.txt":
test_texts = data_helper.get_reviews(os.path.join(DATA_DIR, data_file))
category_texts = {"test data": test_texts}
else:
positive_texts, negative_texts = data_helper.get_reviews(
os.path.join(DATA_DIR, data_file))
category_texts = {
"positive": positive_texts,
"negative": negative_texts
}
# build features
features_category_tuples, texts = get_features_category_tuples(
category_texts, feat_name)
# save features to file
if save_feats is not None:
write_features_category(features_category_tuples, save_feats)
return features_category_tuples, texts