all_B_over_U = data['all_B_over_U']
print 'Done'
N = 500
random_hyperparams = {
'count__ngram_range': [(1, n) for n in npr.randint(1, 4, N)],
'count__max_df': npr.uniform(.75, 1.0, N),
'tfidf__norm': npr.choice(['l1', 'l2', None], N),
'tfidf__use_idf': npr.choice([True, False], N),
'svc__base_estimator__C': npr.choice(np.logspace(-3, 4, 10000), N)
}
score_kwds = {
'xs': train_data['text'],
'ys': train_data['is_multiple'],
'bs': train_data['is_biased'],
'all_B_over_U': all_B_over_U,
'fit_weight_kwd': 'svc__sample_weight',
'n_cv_splits': 5,
'random_seed': random_seed
}
print 'Starting Experiments...'
t0 = time()
experiments = random_search(model, random_hyperparams,
'best_svm_mult_silver.pkl', **score_kwds)
print 'Done {}:{}:{} seconds. Writing out experiments'.format(*hms(time() -
t0))
joblib.dump(experiments, 'svm_mult_silver_dev.pkl')
print 'All done'
print 'Done'
N = 500
random_hyperparams = {
'count__ngram_range': [(1, n) for n in npr.randint(1, 4, N)],
'count__max_df': npr.uniform(.75, 1.0, N),
'tfidf__norm': npr.choice(['l1', 'l2', None], N),
'tfidf__use_idf': npr.choice([True, False], N),
'rf__n_estimators': npr.choice(range(10, 200), N),
'rf__max_features': npr.choice(['log2', 'sqrt'], N)
}
score_kwds = {
'xs': train_data['text'],
'ys': train_data['is_multiple'],
'bs': train_data['is_biased'],
'all_B_over_U': all_B_over_U,
'fit_weight_kwd': 'rf__sample_weight',
'n_cv_splits': 5,
'random_seed': random_seed
}
print 'Starting Experiments...'
t0 = time()
experiments = random_search(model, random_hyperparams, 'best_rf_mult_gold.pkl',
**score_kwds)
print 'Done {}:{}:{} seconds. Writing out experiments'.format(*hms(time() -
t0))
joblib.dump(experiments, 'rf_mult_gold_dev.pkl')
print 'All done'
all_B_over_U = data['all_B_over_U']
print 'Done'
N = 500
random_hyperparams = {
'count__ngram_range': [(1, n) for n in npr.randint(1, 4, N)],
'count__max_df': npr.uniform(.75, 1.0, N),
'tfidf__norm': npr.choice(['l1', 'l2', None], N),
'tfidf__use_idf': npr.choice([True, False], N),
'svc__base_estimator__C': npr.choice(np.logspace(-3, 4, 10000), N)
}
score_kwds = {
'xs': train_data['text'],
'ys': train_data['is_foodborne'],
'bs': train_data['is_biased'],
'all_B_over_U': all_B_over_U,
'fit_weight_kwd': 'svc__sample_weight',
'n_cv_splits': 5,
'random_seed': random_seed
}
print 'Starting Experiments...'
t0 = time()
experiments = random_search(model, random_hyperparams,
'best_svm_sick_biased.pkl', **score_kwds)
print 'Done {}:{}:{} seconds. Writing out experiments'.format(*hms(time() -
t0))
joblib.dump(experiments, 'svm_sick_biased_dev.pkl')
print 'All done'
train_data = data['train_data']
all_B_over_U = data['all_B_over_U']
print 'Done'
N = 500
random_hyperparams = {
'count__ngram_range':[(1,n) for n in npr.randint(1,4, N)],
'count__max_df':npr.uniform(.75, 1.0, N),
'tfidf__norm':npr.choice(['l1', 'l2', None], N),
'tfidf__use_idf':npr.choice([True, False], N),
'logreg__C':npr.choice(np.logspace(-3,4, 10000), N),
'logreg__penalty':npr.choice(['l1', 'l2'], N)
}
score_kwds = {
'xs':train_data['text'],
'ys':train_data['is_foodborne'],
'bs':train_data['is_biased'],
'all_B_over_U':all_B_over_U,
'fit_weight_kwd':'logreg__sample_weight',
'n_cv_splits':5,
'random_seed':random_seed
}
print 'Starting Experiments...'
t0 = time()
experiments = random_search(model, random_hyperparams, 'best_lr_sick_gold.pkl', **score_kwds)
print 'Done {}:{}:{} seconds. Writing out experiments'.format(*hms(time()-t0))
joblib.dump(experiments, 'lr_sick_gold_dev.pkl')
print 'All done'
