def test_max_lines():
sys.argv = ['EvoMSA', '--evodag-kw={"popsize": 10, "early_stopping_rounds": 10, "time_limit": 5, "n_estimators": 5}', '-ot.model', '-n2', TWEETS]
train()
sys.argv = ['EvoMSA', '-mt.model', '--max-lines', '500', '-ot.json', TWEETS]
predict()
os.unlink('t.model')
os.unlink('t.json')
def test_decision_function():
sys.argv = ['EvoMSA', '--evodag-kw={"popsize": 10, "early_stopping_rounds": 10, "time_limit": 5, "n_estimators": 5}',
'-ot.model', '-n2', TWEETS, TWEETS]
train(output=True)
sys.argv = ['EvoMSA', '--decision-function', '-mt.model', '-ot1.json', TWEETS]
predict()
df = [x['decision_function'] for x in tweet_iterator('t1.json')]
assert len(df[0]) == 4
os.unlink('t1.json')
os.unlink('t.model')
def test_raw_outputs():
sys.argv = ['EvoMSA', '--kw={"stacked_method_args": {"popsize": 10, "early_stopping_rounds": 10, "time_limit": 60, "n_estimators": 30}}',
'-ot.model', '-n2', TWEETS]
train(output=True)
sys.argv = ['EvoMSA', '--raw-outputs', '-mt.model', '-ot1.json', TWEETS]
predict()
df = [x['decision_function'] for x in tweet_iterator('t1.json')]
assert len(df[0]) == 30 * 4
os.unlink('t1.json')
os.unlink('t.model')
def test_predict():
import numpy as np
sys.argv = ['EvoMSA', '--evodag-kw={"popsize": 10, "early_stopping_rounds": 10, "time_limit": 5, "n_estimators": 5}',
'-ot.model', '-n2', TWEETS, TWEETS]
train(output=True)
sys.argv = ['EvoMSA', '-mt.model', '-ot1.json', TWEETS]
predict()
hy = np.array([x['klass'] for x in tweet_iterator('t1.json')])
[x['decision_function'] for x in tweet_iterator('t1.json')]
y = np.array([x['klass'] for x in tweet_iterator(TWEETS)])
acc = (y == hy).mean()
print(acc)
assert acc <= 1 and acc > 0.8
os.unlink('t1.json')
os.unlink('t.model')
def test_predict_numbers():
from sklearn.preprocessing import LabelEncoder
import json
d = [x for x in tweet_iterator(TWEETS)]
le = LabelEncoder().fit([x['klass'] for x in d])
y = le.transform([x['klass'] for x in d]).tolist()
with open('ex.json', 'w') as fpt:
for x, y0 in zip(d, y):
x['klass'] = y0
fpt.write(json.dumps(x) + '\n')
sys.argv = ['EvoMSA', '--kw={"stacked_method_args": {"popsize": 10, "early_stopping_rounds": 10, "time_limit": 5, "n_estimators": 5}, "models": [["EvoMSA.model.Corpus", "EvoMSA.model.Bernoulli"]], "TR": false}',
'-ot.model', '-n1', 'ex.json']
train(output=True)
sys.argv = ['EvoMSA', '-mt.model', '-ot1.json', TWEETS]
predict()
def test_predict_NearestCentroid():
from sklearn.preprocessing import LabelEncoder
import json
d = [x for x in tweet_iterator(TWEETS)]
le = LabelEncoder().fit([x['klass'] for x in d])
y = le.transform([x['klass'] for x in d]).tolist()
with open('ex.json', 'w') as fpt:
for x, y0 in zip(d, y):
x['klass'] = y0
fpt.write(json.dumps(x) + '\n')
sys.argv = ['EvoMSA',
'--kw={"evodag_class": "sklearn.neighbors.NearestCentroid", "TR": false, "models": [["EvoMSA.model.Corpus", "EvoMSA.model.Bernulli"]]}',
'-ot.model', '-n1', 'ex.json']
train(output=True)
sys.argv = ['EvoMSA', '-mt.model', '-ot1.json', TWEETS]
predict()
def test_performance_public_set():
import os
from EvoMSA.command_line import performance
for seed in range(4):
if os.path.isfile('t-%s.model' % seed):
continue
sys.argv = ['EvoMSA', '--evodag-kw={"popsize": 10, "early_stopping_rounds": 10, "time_limit": 5, "n_estimators": 3}',
'--kw={"seed": %s}' % seed, '-ot-%s.model' % seed, '-n1', TWEETS]
train(output=True)
for seed in range(4):
if os.path.isfile('t-%s.predict' % seed):
continue
sys.argv = ['EvoMSA', '-mt-%s.model' % seed, '-ot-%s.predict' % seed, TWEETS]
predict()
for score in ['macroF1', 'macroRecall', 'macroPrecision', 'accuracy']:
sys.argv = ['EvoMSA', '--score', score, '-n2', '-y', TWEETS] + ['t-%s.predict' % seed for seed in range(2)] + ['-'] + ['t-%s.predict' % seed for seed in range(2, 4)]
m = performance(output=True)
assert len(m._p) == 2