def test_acc():
presented = [[['cat', 'bat', 'hat', 'goat'],
['zoo', 'animal', 'zebra', 'horse']]]
recalled = [[['bat', 'cat', 'goat', 'hat'], ['animal', 'horse', 'zoo']]]
egg = Egg(pres=presented, rec=recalled)
assert np.allclose(
egg.analyze('accuracy').data.values,
[np.array([1.]), np.array([.75])])
def test_pfr_smooth_correlation_3d():
presented = [[[[10, 0, 10], [20, 0, 0], [30, 0, -10], [40, 0, -20]],
[[10, 0, 10], [20, 0, 0], [30, 0, -10], [40, 0, -20]]]]
recalled = [[[[20, 0, 0], [10, 0, 10], [40, 0, -20], [30, 0, -10]],
[[20, 0, 0], [40, 0, -20], [10, 0, 10]]]]
egg = Egg(pres=presented, rec=recalled)
egg.analyze('pfr', match='smooth', distance='correlation',
features='item').data.values
def test_spc_best_euclidean():
presented = [[[10, 20, 30, 40], [10, 20, 30, 40]]]
recalled = [[[20, 10, 40, 30], [20, 40, 10]]]
egg = Egg(pres=presented, rec=recalled)
assert np.allclose(
egg.analyze('spc', match='best', distance='euclidean',
features='item').data.values,
[np.array([1., 1., 1., 1.]),
np.array([1., 1., 0., 1.])])
def test_analysis_pfr():
presented = [[['cat', 'bat', 'hat', 'goat'],
['zoo', 'animal', 'zebra', 'horse']]]
recalled = [[['bat', 'cat', 'goat', 'hat'], ['animal', 'horse', 'zoo']]]
egg = Egg(pres=presented, rec=recalled)
assert np.array_equal(
egg.analyze('pfr').data.values,
[np.array([0., 1., 0., 0.]),
np.array([0., 1., 0., 0.])])
def test_pfr_best_euclidean_3d_exception_no_features():
presented = [[[[10, 0, 0], [20, 0, 0], [30, 0, 0], [40, 0, 0]],
[[10, 0, 0], [20, 0, 0], [30, 0, 0], [40, 0, 0]]]]
recalled = [[[[20, 0, 0], [10, 0, 0], [40, 0, 0], [30, 0, 0]],
[[20, 0, 0], [40, 0, 0], [10, 0, 0]]]]
egg = Egg(pres=presented, rec=recalled)
with pytest.raises(Exception):
assert np.array_equal(
egg.analyze('pfr', match='best', distance='euclidean').data.values,
[np.array([0., 1., 0., 0.]),
np.array([0., 1., 0., 0.])])
def test_pfr_best_euclidean_3d_exception_item_specified():
presented = [[[[10, 0, 0], [20, 0, 0], [30, 0, 0], [40, 0, 0]],
[[10, 0, 0], [20, 0, 0], [30, 0, 0], [40, 0, 0]]]]
recalled = [[[[20, 0, 0], [10, 0, 0], [40, 0, 0], [30, 0, 0]],
[[20, 0, 0], [40, 0, 0], [10, 0, 0]]]]
egg = Egg(pres=presented, rec=recalled)
assert np.array_equal(
egg.analyze('pfr', match='best', distance='euclidean',
features='item').data.values,
[np.array([0., 1., 0., 0.]),
np.array([0., 1., 0., 0.])])
def test_lagcrp_exact():
# example from kahana lab lag-crp tutorial
presented = [[['1', '2', '3', '4', '5', '6', '7', '8']]]
recalled = [[['8', '7', '1', '2', '3', '5', '6', '4']]]
egg = Egg(pres=presented, rec=recalled)
assert np.allclose(egg.analyze('lagcrp').data.values,
np.array([[
0.0, 0.0, 0.5, 0.0, 0.0, 0.0, 0.333333, 0.333333,
np.nan, 0.75, 0.333333, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0
]]),
equal_nan=True)
def test_analysis_acc_multisubj():
presented = [[['cat', 'bat', 'hat', 'goat'],
['zoo', 'animal', 'zebra', 'horse']],
[['cat', 'bat', 'hat', 'goat'],
['zoo', 'animal', 'zebra', 'horse']]]
recalled = [[['bat', 'cat', 'goat', 'hat'], ['animal', 'horse', 'zoo']],
[['bat', 'cat', 'goat', 'hat'], ['animal', 'horse', 'zoo']]]
multisubj_egg = Egg(pres=presented, rec=recalled)
assert np.array_equal(
multisubj_egg.analyze('accuracy').data.values,
np.array([[1.], [.75], [1.], [.75]]))
def test_analysis_spc_multisubj():
presented = [[['cat', 'bat', 'hat', 'goat'],
['zoo', 'animal', 'zebra', 'horse']],
[['cat', 'bat', 'hat', 'goat'],
['zoo', 'animal', 'zebra', 'horse']]]
recalled = [[['bat', 'cat', 'goat', 'hat'], ['animal', 'horse', 'zoo']],
[['bat', 'cat', 'goat', 'hat'], ['animal', 'horse', 'zoo']]]
multisubj_egg = Egg(pres=presented, rec=recalled)
assert np.allclose(
multisubj_egg.analyze('spc').data.values,
np.array([[1., 1., 1., 1.], [1., 1., 0., 1.], [1., 1., 1., 1.],
[1., 1., 0., 1.]]))
def test_acc_best_correlation_3d():
presented = [[[[10, 0, 10], [20, 0, 0], [30, 0, -10], [40, 0, -20]],
[[10, 0, 10], [20, 0, 0], [30, 0, -10], [40, 0, -20]]]]
recalled = [[[[20, 0, 0], [10, 0, 10], [40, 0, -20], [30, 0, -10]],
[[20, 0, 0], [40, 0, -20], [10, 0, 10]]]]
egg = Egg(pres=presented, rec=recalled)
assert np.array_equal(
egg.analyze('accuracy',
match='best',
distance='correlation',
features='item').data.values,
[np.array([1.]), np.array([.75])])
def test_lagcrp_exact_3d():
presented = [[[[10, 0, 0], [20, 0, 0], [30, 0, 0], [40, 0, 0], [50, 0, 0],
[60, 0, 0], [70, 0, 0], [80, 0, 0]]]]
recalled = [[[[80, 0, 0], [70, 0, 0], [10, 0, 0], [20, 0, 0], [30, 0, 0],
[50, 0, 0], [60, 0, 0], [40, 0, 0]]]]
egg = Egg(pres=presented, rec=recalled)
assert np.allclose(egg.analyze('lagcrp', match='best',
features='item').data.values,
np.array([[
0.0, 0.0, 0.5, 0.0, 0.0, 0.0, 0.333333, 0.333333,
np.nan, 0.75, 0.333333, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0
]]),
equal_nan=True)
def test_lagcrp_best_euclidean():
presented = [[[10, 20, 30, 40, 50, 60, 70, 80]]]
recalled = [[[81, 71, 11, 21, 31, 51, 61, 41]]]
egg = Egg(pres=presented, rec=recalled)
assert np.allclose(egg.analyze('lagcrp',
match='best',
distance='euclidean',
features='item').data.values,
np.array([[
0.0, 0.0, 0.5, 0.0, 0.0, 0.0, 0.333333, 0.333333,
np.nan, 0.75, 0.333333, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0
]]),
equal_nan=True)
def test_acc_best_euclidean_3d_features_not_set():
presented = [[[{
'item': i,
'feature1': [i * 10, 0, 0]
} for i in range(1, 5)] for i in range(2)]]
recalled = [[[{
'item': i,
'feature1': [i * 10, 0, 0]
} for i in [2, 1, 4, 3]],
[{
'item': i,
'feature1': [i * 10, 0, 0]
} for i in [2, 4, 1]]]]
egg = Egg(pres=presented, rec=recalled)
assert np.array_equal(
egg.analyze('pfr', match='best', distance='euclidean').data.values,
[np.array([0., 1., 0., 0.]),
np.array([0., 1., 0., 0.])])
def test_acc_best_euclidean():
presented = [[[{
'item': i,
'feature1': i * 10
} for i in range(1, 5)] for i in range(2)]]
recalled = [[[{
'item': i,
'feature1': i * 10
} for i in [2, 1, 4, 3]],
[{
'item': i,
'feature1': i * 10
} for i in [2, 4, 1]]]]
egg = Egg(pres=presented, rec=recalled)
assert np.array_equal(
egg.analyze('accuracy',
match='best',
distance='euclidean',
features=['feature1']).data.values,
[np.array([1.]), np.array([.75])])
def test_spc_best_euclidean():
presented = [[[{
'item': i,
'feature1': i * 10
} for i in range(1, 5)] for i in range(2)]]
recalled = [[[{
'item': i,
'feature1': i * 10
} for i in [2, 1, 4, 3]],
[{
'item': i,
'feature1': i * 10
} for i in [2, 4, 1]]]]
egg = Egg(pres=presented, rec=recalled)
assert np.allclose(
egg.analyze('spc',
match='best',
distance='euclidean',
features='feature1').data.values,
[np.array([1., 1., 1., 1.]),
np.array([1., 1., 0., 1.])])
def test_spc_best_euclidean_3d_2features():
presented = [[[{
'item': i,
'feature1': [i * 10, 0, 0],
'feature2': [i * 10, 0, 0]
} for i in range(1, 5)] for i in range(2)]]
recalled = [[[{
'item': i,
'feature1': [i * 10, 0, 0],
'feature2': [i * 10, 0, 0]
} for i in [2, 1, 4, 3]],
[{
'item': i,
'feature1': [i * 10, 0, 0],
'feature2': [i * 10, 0, 0]
} for i in [2, 4, 1]]]]
egg = Egg(pres=presented, rec=recalled)
assert np.array_equal(
egg.analyze('spc',
match='best',
distance='euclidean',
features=['feature1', 'feature2']).data.values,
[np.array([1., 1., 1., 1.]),
np.array([1., 1., 0., 1.])])
# -*- coding: utf-8 -*-
from quail.analysis import analyze
from quail.load import load_example_data
from quail.egg import Egg
import numpy as np
import pytest
import pandas as pd
presented = [[['cat', 'bat', 'hat', 'goat'],
['zoo', 'animal', 'zebra', 'horse']]]
recalled = [[['bat', 'cat', 'goat', 'hat'], ['animal', 'horse', 'zoo']]]
egg = Egg(pres=presented, rec=recalled)
def test_analysis_acc():
print(analyze(egg, analysis='accuracy'))
print([np.array([1.]), np.array([.75])])
assert np.array_equal(
analyze(egg, analysis='accuracy').data.values,
[np.array([1.]), np.array([.75])])
def test_analysis_spc():
assert np.array_equal(
analyze(egg, analysis='spc').data.values,
[np.array([1., 1., 1., 1.]),
np.array([1., 1., 0., 1.])])
def test_analysis_spc_listgroup():
'category' : 'animal',
'size' : 'bigger',
'starting letter' : 'D',
'length' : 3
},
'SHOE' : {
'category' : 'object',
'size' : 'smaller',
'starting letter' : 'S',
'length' : 4
},
'HORSE' : {
'category' : 'animal',
'size' : 'bigger',
'starting letter' : 'H',
'length' : 5
}}
presented = list(map(lambda x: {'item' : x}, presented))
for p in presented:
p.update(**features[p['item']])
recalled = list(map(lambda x: {'item' : x}, recalled))
for r in recalled:
r.update(**features[r['item']])
egg = Egg(pres=[[presented]],rec=[[recalled]])
def test_temporal():
egg.analyze('temporal')
def test_egg():
list1 = [[[1, 2], [3, 4]], [[5, 6], [7, 8]]]
list2 = [[[10, 20], [30, 40]], [[50, 60], [70, 80]]]
egg = Egg(pres=list1, rec=list2)
assert type(egg.pres) == pd.core.frame.DataFrame
assert type(egg.rec) == pd.core.frame.DataFrame
'size': 'bigger',
'starting letter': 'D',
'length': 3
}, {
'category': 'object',
'size': 'smaller',
'starting letter': 'S',
'length': 4
}, {
'category': 'animal',
'size': 'bigger',
'starting letter': 'H',
'length': 5
}]
egg = Egg(pres=[presented], rec=[recalled], features=[features])
def test_egg_works():
assert isinstance(egg, Egg)
def test_pres_is_df():
assert isinstance(egg.pres, pd.DataFrame)
def test_rec_is_df():
assert isinstance(egg.rec, pd.DataFrame)
def test_pres_cells_are_dicts():
def test_egg_recmat():
recmat = [[[3, 4, 2, 1, 0], [4, 2, 5, 3, 2]]]
egg = Egg(recmat=recmat, list_length=10)
assert isinstance(egg, Egg)
# -*- coding: utf-8 -*-
from quail.egg import Egg, FriedEgg
import pytest
import pandas as pd
import six
import matplotlib.pyplot as plt
presented = [[['cat', 'bat', 'hat', 'goat'],
['zoo', 'animal', 'zebra', 'horse']]]
recalled = [[['bat', 'cat', 'goat', 'hat'], ['animal', 'horse', 'zoo']]]
fried_egg = Egg(pres=presented, rec=recalled).analyze('accuracy')
def test_fried_egg_data_is_df():
assert isinstance(fried_egg.data, pd.DataFrame)
def test_fried_egg_analysis():
assert isinstance(fried_egg.analysis, six.string_types)
assert fried_egg.analysis == 'accuracy'
def test_fried_egg_listlength():
assert fried_egg.list_length == 4
def test_fried_egg_n_lists():
assert fried_egg.n_lists == 2
