parser = argparse.ArgumentParser(
description='Run all queries on the inverted index.')
parser.add_argument(
'--new',
default=True,
help='If True then build a new index from scratch. If False then attempt to'
' reuse existing index')
parser.add_argument(
'--sim',
default='BM25',
help='The type of similarity to use. Should be "TF" or "TFIDF" or "BM25"')
args = parser.parse_args()
index = InvertedIndex(Preprocessor())
index.index_directory(os.path.join('gov', 'documents'),
use_stored_index=(not args.new))
sim_name_to_class = {
'TF': TF_Similarity,
'TFIDF': TFIDF_Similarity,
'BM25': BM25_Similarity
}
sim = sim_name_to_class[args.sim]
index.set_similarity(sim)
print(f'Setting similarity to {sim.__name__}')
print()
print('Index ready.')
topics_file = os.path.join('gov', 'topics', 'gov.topics')
import argparse
import os
from inverted_index import InvertedIndex
from preprocessor import Preprocessor
from similarity_measures import TF_Similarity, TFIDF_Similarity, BM25_Similarity
parser = argparse.ArgumentParser(description='Run all queries on the inverted index.')
parser.add_argument('--new', default=True, help='If True then build a new index from scratch. If False then attempt to'
' reuse existing index')
parser.add_argument('--sim', default='BM25', help='The type of similarity to use. Should be "TF" or "TFIDF" or "BM25')
args = parser.parse_args()
index = InvertedIndex(Preprocessor())
index.index_directory(os.path.join('gov', 'documents'), use_stored_index=True)
sim_name_to_class = {'TF': TF_Similarity,
'TFIDF': TFIDF_Similarity,
'BM25': BM25_Similarity}
sim = sim_name_to_class[args.sim]
index.set_similarity(sim)
print(f'Setting similarity to {sim.__name__}')
print()
print('Index ready.')
topics_file = os.path.join('gov', 'topics', 'gov.topics')
runs_file = os.path.join('runs', 'retrieved.runs')