def feature_consolidation(resume_text, top_unigram_list, top_bigram_list):
"""
Function to consolidate all the featuresets for the training data
Args:
top_unigram_list -- list of top unigrams from the training dataset
top_bigram_list -- list of top bigrams from the training dataset
Returns:
consolidated_features -- list of consolidated features
"""
uni_feats = [unigram_features(resume_text, top_unigram_list)]
bi_feats = [bigram_features(resume_text, top_bigram_list)]
consolidated_features = []
ind = 0
while ind < len(uni_feats):
consolidated_features.append(uni_feats[ind] + bi_feats[ind])
ind += 1
return consolidated_features
def feature_consolidation(resume_text, top_unigram_list, top_bigram_list):
"""
Function to consolidate all the featuresets for the training data
Args:
top_unigram_list -- list of top unigrams from the training dataset
top_bigram_list -- list of top bigrams from the training dataset
Returns:
consolidated_features -- list of consolidated features
"""
uni_feats = [unigram_features(resume_text, top_unigram_list)]
bi_feats = [bigram_features(resume_text, top_bigram_list)]
consolidated_features = []
ind = 0
while ind < len(uni_feats):
consolidated_features.append(uni_feats[ind] + bi_feats[ind])
ind += 1
return consolidated_features
def feature_consolidation(resumes, top_unigram_list, top_bigram_list, add_true_score=False):
"""
Function to consolidate all the featuresets for the training data
Args:
resumes -- list of tuples [(resume_text, tag, filename), (resume_text, tag, filename)...]
top_unigram_list -- list of top unigrams from the training dataset
top_bigram_list -- list of top bigrams from the training dataset
add_true_score -- boolean (default: False)
Returns:
consolidated_features -- list of consolidated features
"""
uni_feats = [unigram_features(resume_text, top_unigram_list) for (resume_text, label, fname) in resumes]
bi_feats = [bigram_features(resume_text, top_bigram_list) for (resume_text, label, fname) in resumes]
consolidated_features = []
ind = 0
while ind < len(uni_feats):
consolidated_features.append(uni_feats[ind] + bi_feats[ind])
ind += 1
return consolidated_features
def test_should_return_zero_if_bigram_is_not_present_in_top_bigrams():
bigram_feats = bigram_features(sample_failing_text, top_bigrams)
for bi_feat in bigram_feats:
assert_equals(0, bi_feat)
def test_should_return_one_if_bigram_is_present_in_top_bigrams():
bigram_feats = bigram_features(sample_passing_text, top_bigrams)
for bi_feat in bigram_feats:
assert_equals(1, bi_feat)
def test_lengths_of_bigram_features_and_top_bigrams_should_be_equal():
bigram_feats = bigram_features(sample_passing_text, top_bigrams)
assert_equals(len(bigram_feats), len(top_bigrams))
def test_bigram_features_for_a_given_text_should_not_be_empty():
bigram_feats = bigram_features(sample_passing_text, top_bigrams)
assert_true(len(bigram_feats))
def test_should_return_zero_if_bigram_is_not_present_in_top_bigrams():
bigram_feats = bigram_features(sample_failing_text, top_bigrams)
for bi_feat in bigram_feats:
assert_equals(0, bi_feat)
def test_should_return_one_if_bigram_is_present_in_top_bigrams():
bigram_feats = bigram_features(sample_passing_text, top_bigrams)
for bi_feat in bigram_feats:
assert_equals(1, bi_feat)
def test_lengths_of_bigram_features_and_top_bigrams_should_be_equal():
bigram_feats = bigram_features(sample_passing_text, top_bigrams)
assert_equals(len(bigram_feats), len(top_bigrams))
def test_bigram_features_for_a_given_text_should_not_be_empty():
bigram_feats = bigram_features(sample_passing_text, top_bigrams)
assert_true(len(bigram_feats))
