Comparación entre FeatureHasher y DictVectorizer

Compara FeatureHasher y DictVectorizer utilizando ambos para vectorizar documentos de texto.

El ejemplo sólo demuestra la sintaxis y la velocidad; en realidad no hace nada útil con los vectores extraídos. Ver los scripts de ejemplo {document_classification_20newsgroups,clustering}.py para el aprendizaje real en documentos de texto.

Es de esperar que haya una discrepancia entre el número de términos reportados por DictVectorizer y por FeatureHasher debido a las colisiones de hash.

Out:

Usage: /home/mapologo/Descargas/scikit-learn-0.24.X/examples/text/plot_hashing_vs_dict_vectorizer.py [n_features_for_hashing]
    The default number of features is 2**18.

Loading 20 newsgroups training data
3803 documents - 6.245MB

DictVectorizer
done in 1.173311s at 5.322MB/s
Found 47928 unique terms

FeatureHasher on frequency dicts
done in 0.705522s at 8.851MB/s
Found 43873 unique terms

FeatureHasher on raw tokens
done in 0.657561s at 9.497MB/s
Found 43873 unique terms

# Author: Lars Buitinck
# License: BSD 3 clause
from collections import defaultdict
import re
import sys
from time import time

import numpy as np

from sklearn.datasets import fetch_20newsgroups
from sklearn.feature_extraction import DictVectorizer, FeatureHasher


def n_nonzero_columns(X):
    """Returns the number of non-zero columns in a CSR matrix X."""
    return len(np.unique(X.nonzero()[1]))


def tokens(doc):
    """Extract tokens from doc.

    This uses a simple regex to break strings into tokens. For a more
    principled approach, see CountVectorizer or TfidfVectorizer.
    """
    return (tok.lower() for tok in re.findall(r"\w+", doc))


def token_freqs(doc):
    """Extract a dict mapping tokens from doc to their frequencies."""
    freq = defaultdict(int)
    for tok in tokens(doc):
        freq[tok] += 1
    return freq


categories = [
    'alt.atheism',
    'comp.graphics',
    'comp.sys.ibm.pc.hardware',
    'misc.forsale',
    'rec.autos',
    'sci.space',
    'talk.religion.misc',
]
# Uncomment the following line to use a larger set (11k+ documents)
# categories = None

print(__doc__)
print("Usage: %s [n_features_for_hashing]" % sys.argv[0])
print("    The default number of features is 2**18.")
print()

try:
    n_features = int(sys.argv[1])
except IndexError:
    n_features = 2 ** 18
except ValueError:
    print("not a valid number of features: %r" % sys.argv[1])
    sys.exit(1)


print("Loading 20 newsgroups training data")
raw_data, _ = fetch_20newsgroups(subset='train', categories=categories,
                                 return_X_y=True)
data_size_mb = sum(len(s.encode('utf-8')) for s in raw_data) / 1e6
print("%d documents - %0.3fMB" % (len(raw_data), data_size_mb))
print()

print("DictVectorizer")
t0 = time()
vectorizer = DictVectorizer()
vectorizer.fit_transform(token_freqs(d) for d in raw_data)
duration = time() - t0
print("done in %fs at %0.3fMB/s" % (duration, data_size_mb / duration))
print("Found %d unique terms" % len(vectorizer.get_feature_names()))
print()

print("FeatureHasher on frequency dicts")
t0 = time()
hasher = FeatureHasher(n_features=n_features)
X = hasher.transform(token_freqs(d) for d in raw_data)
duration = time() - t0
print("done in %fs at %0.3fMB/s" % (duration, data_size_mb / duration))
print("Found %d unique terms" % n_nonzero_columns(X))
print()

print("FeatureHasher on raw tokens")
t0 = time()
hasher = FeatureHasher(n_features=n_features, input_type="string")
X = hasher.transform(tokens(d) for d in raw_data)
duration = time() - t0
print("done in %fs at %0.3fMB/s" % (duration, data_size_mb / duration))
print("Found %d unique terms" % n_nonzero_columns(X))