#!/usr/bin/env python # -*- coding: utf-8 -*- # # Copyright (C) 2011 Radim Rehurek # Licensed under the GNU LGPL v2.1 - http://www.gnu.org/licenses/lgpl.html """ USAGE: %(program)s CORPUS_DENSE.mm CORPUS_SPARSE.mm [NUMDOCS] Run speed test of similarity queries. Only use the first NUMDOCS documents of \ each corpus for testing (or use all if no NUMDOCS is given). The two sample corpora can be downloaded from http://nlp.fi.muni.cz/projekty/gensim/wikismall.tgz Example: ./simspeed2.py wikismall.dense.mm wikismall.sparse.mm """ import logging import sys import itertools import os import math from time import time import gensim if __name__ == '__main__': logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO) logging.info("running %s", " ".join(sys.argv)) # check and process cmdline input program = os.path.basename(sys.argv[0]) if len(sys.argv) < 3: print(globals()['__doc__'] % locals()) sys.exit(1) corpus_dense = gensim.corpora.MmCorpus(sys.argv[1]) corpus_sparse = gensim.corpora.MmCorpus(sys.argv[2]) dense_features, sparse_features = corpus_dense.num_terms, corpus_sparse.num_terms if len(sys.argv) > 3: NUMDOCS = int(sys.argv[3]) corpus_dense = list(itertools.islice(corpus_dense, NUMDOCS)) corpus_sparse = list(itertools.islice(corpus_sparse, NUMDOCS)) # create the query index to be tested (one for dense input, one for sparse) index_dense = gensim.similarities.Similarity('/tmp/tstdense', corpus_dense, dense_features) index_sparse = gensim.similarities.Similarity('/tmp/tstsparse', corpus_sparse, sparse_features) density = 100.0 * sum(shard.num_nnz for shard in index_sparse.shards) / (len(index_sparse) * sparse_features) logging.info( "test 1 (dense): similarity of all vs. all (%i documents, %i dense features)", len(corpus_dense), index_dense.num_features ) for chunksize in [1, 8, 32, 64, 128, 256, 512, 1024, index_dense.shardsize]: index_dense.chunksize = chunksize start = time() for sim in index_dense: pass taken = time() - start queries = math.ceil(1.0 * len(corpus_dense) / chunksize) logging.info( "chunksize=%i, time=%.4fs (%.2f docs/s, %.2f queries/s)", chunksize, taken, len(corpus_dense) / taken, queries / taken ) index_dense.num_best = 10 logging.info("test 2 (dense): as above, but only ask for the top-10 most similar for each document") for chunksize in [1, 8, 32, 64, 128, 256, 512, 1024, index_dense.shardsize]: index_dense.chunksize = chunksize start = time() sims = [sim for sim in index_dense] taken = time() - start queries = math.ceil(1.0 * len(corpus_dense) / chunksize) logging.info( "chunksize=%i, time=%.4fs (%.2f docs/s, %.2f queries/s)", chunksize, taken, len(corpus_dense) / taken, queries / taken ) index_dense.num_best = None logging.info( "test 3 (sparse): similarity of all vs. all (%i documents, %i features, %.2f%% density)", len(corpus_sparse), index_sparse.num_features, density ) for chunksize in [1, 5, 10, 100, 256, 500, 1000, index_sparse.shardsize]: index_sparse.chunksize = chunksize start = time() for sim in index_sparse: pass taken = time() - start queries = math.ceil(1.0 * len(corpus_sparse) / chunksize) logging.info( "chunksize=%i, time=%.4fs (%.2f docs/s, %.2f queries/s)", chunksize, taken, len(corpus_sparse) / taken, queries / taken ) index_sparse.num_best = 10 logging.info("test 4 (sparse): as above, but only ask for the top-10 most similar for each document") for chunksize in [1, 5, 10, 100, 256, 500, 1000, index_sparse.shardsize]: index_sparse.chunksize = chunksize start = time() for sim in index_sparse: pass taken = time() - start queries = math.ceil(1.0 * len(corpus_sparse) / chunksize) logging.info( "chunksize=%i, time=%.4fs (%.2f docs/s, %.2f queries/s)", chunksize, taken, len(corpus_sparse) / taken, queries / taken ) index_sparse.num_best = None # Difference between test #5 and test #1 is that the query in #5 is a gensim iterable # corpus, while in #1, the index is used directly (numpy arrays). So #5 is slower, # because it needs to convert sparse vecs to numpy arrays and normalize them to # unit length=extra work, which #1 avoids. query = list(itertools.islice(corpus_dense, 1000)) logging.info( "test 5 (dense): dense corpus of %i docs vs. index (%i documents, %i dense features)", len(query), len(index_dense), index_dense.num_features ) for chunksize in [1, 8, 32, 64, 128, 256, 512, 1024]: start = time() if chunksize > 1: sims = [] for chunk in gensim.utils.chunkize_serial(query, chunksize): _ = index_dense[chunk] else: for vec in query: _ = index_dense[vec] taken = time() - start queries = math.ceil(1.0 * len(query) / chunksize) logging.info( "chunksize=%i, time=%.4fs (%.2f docs/s, %.2f queries/s)", chunksize, taken, len(query) / taken, queries / taken ) # Same comment as for test #5. query = list(itertools.islice(corpus_dense, 1000)) logging.info( "test 6 (sparse): sparse corpus of %i docs vs. sparse index (%i documents, %i features, %.2f%% density)", len(query), len(corpus_sparse), index_sparse.num_features, density ) for chunksize in [1, 5, 10, 100, 500, 1000]: start = time() if chunksize > 1: sims = [] for chunk in gensim.utils.chunkize_serial(query, chunksize): _ = index_sparse[chunk] else: for vec in query: _ = index_sparse[vec] taken = time() - start queries = math.ceil(1.0 * len(query) / chunksize) logging.info( "chunksize=%i, time=%.4fs (%.2f docs/s, %.2f queries/s)", chunksize, taken, len(query) / taken, queries / taken ) logging.info("finished running %s", program)