1
0
Fork 0
QuaPy/LeQua2022/baselines.py

109 lines
4.4 KiB
Python

import argparse
import pickle
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.linear_model import LogisticRegression as LR
from quapy.method.aggregative import *
from quapy.method.non_aggregative import MaximumLikelihoodPrevalenceEstimation as MLPE
import quapy.functional as F
from data import *
import os
import constants
# LeQua official baselines for task T1A (Binary/Vector) and T1B (Multiclass/Vector)
# =========================================================
def baselines():
yield CC(LR(n_jobs=-1)), "CC"
# yield ACC(LR(n_jobs=-1)), "ACC"
# yield PCC(LR(n_jobs=-1)), "PCC"
yield PACC(LR(n_jobs=-1)), "PACC"
yield EMQ(CalibratedClassifierCV(LR(), n_jobs=-1)), "SLD"
# yield HDy(LR(n_jobs=-1)) if args.task == 'T1A' else OneVsAll(HDy(LR()), n_jobs=-1), "HDy"
# yield MLPE(), "MLPE"
def main(args):
models_path = qp.util.create_if_not_exist(os.path.join(args.modeldir, args.task))
path_dev_vectors = os.path.join(args.datadir, 'dev_samples')
path_dev_prevs = os.path.join(args.datadir, 'dev_prevalences.txt')
path_train = os.path.join(args.datadir, 'training_data.txt')
qp.environ['SAMPLE_SIZE'] = constants.SAMPLE_SIZE[args.task]
if args.task in {'T1A', 'T1B'}:
train = LabelledCollection.load(path_train, load_vector_documents)
def gen_samples():
return gen_load_samples(path_dev_vectors, ground_truth_path=path_dev_prevs, load_fn=load_vector_documents)
else:
train = LabelledCollection.load(path_train, load_raw_documents)
tfidf = TfidfVectorizer(min_df=5, sublinear_tf=True, ngram_range=(1, 2))
train.instances = tfidf.fit_transform(*train.Xy)
def gen_samples():
return gen_load_samples(path_dev_vectors, ground_truth_path=path_dev_prevs,
load_fn=load_raw_documents, vectorizer=tfidf)
print(f'number of classes: {len(train.classes_)}')
print(f'number of training documents: {len(train)}')
print(f'training prevalence: {F.strprev(train.prevalence())}')
print(f'training matrix shape: {train.instances.shape}')
param_grid = {
'C': np.logspace(-3, 3, 7),
'class_weight': ['balanced', None]
}
param_grid = {
'C': [0.01],
'class_weight': ['balanced']
}
for quantifier, q_name in baselines():
print(f'{q_name}: Model selection')
quantifier = qp.model_selection.GridSearchQ(
quantifier,
param_grid,
sample_size=None,
protocol='gen',
error=qp.error.mrae,
refit=False,
verbose=True
).fit(train, gen_samples)
print(f'{q_name} got MRAE={quantifier.best_score_:.5f} (hyper-params: {quantifier.best_params_})')
model_path = os.path.join(models_path, q_name+'.pkl')
print(f'saving model in {model_path}')
pickle.dump(quantifier.best_model(), open(model_path, 'wb'), protocol=pickle.HIGHEST_PROTOCOL)
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='LeQua2022 baselines')
parser.add_argument('task', metavar='TASK', type=str, choices=['T1A', 'T1B', 'T2A', 'T2B'],
help='Task name (T1A, T1B, T2A, T2B)')
parser.add_argument('datadir', metavar='DATA-PATH', type=str,
help='Path of the directory containing "dev_prevalences.txt", "training_data.txt", and '
'the directory "dev_samples"')
parser.add_argument('modeldir', metavar='MODEL-PATH', type=str,
help='Path where to save the models. '
'A subdirectory named <task> will be automatically created.')
args = parser.parse_args()
if not os.path.exists(args.datadir):
raise FileNotFoundError(f'path {args.datadir} does not exist')
if not os.path.isdir(args.datadir):
raise ValueError(f'path {args.datadir} is not a valid directory')
if not os.path.exists(os.path.join(args.datadir, "dev_prevalences.txt")):
raise FileNotFoundError(f'path {args.datadir} does not contain "dev_prevalences.txt" file')
if not os.path.exists(os.path.join(args.datadir, "training_data.txt")):
raise FileNotFoundError(f'path {args.datadir} does not contain "training_data.txt" file')
if not os.path.exists(os.path.join(args.datadir, "dev_samples")):
raise FileNotFoundError(f'path {args.datadir} does not contain "dev_samples" folder')
main(args)