tonegas/nnodely | Build 14359872492 | tests/test_visualizer.py | Coveralls

1

import sys, os, unittest, torch, shutil

1✔

2

import numpy as np

1✔

4

from nnodely import *

1✔

5

from nnodely.basic.relation import NeuObj

1✔

6

from nnodely.support.logger import logging, nnLogger

1✔

8

log = nnLogger(__name__, logging.CRITICAL)

1✔

9

log.setAllLevel(logging.CRITICAL)

1✔

11

sys.path.append(os.getcwd())

1✔

16

class ModelyTestVisualizer(unittest.TestCase):

1✔

17

    def __init__(self, *args, **kwargs):

1✔

18

        NeuObj.clearNames()

1✔

19

        super(ModelyTestVisualizer, self).__init__(*args, **kwargs)

1✔

21

        self.x = x = Input('x')

1✔

22

        self.y = y = Input('y')

1✔

23

        self.z = z = Input('z')

1✔

26

        def myFun(K1, p1, p2):

1✔

27

            return K1 * p1 * p2

1✔

29

        K_x = Parameter('k_x', dimensions=1, tw=1, init='init_constant', init_params={'value': 1})

1✔

30

        K_y = Parameter('k_y', dimensions=1, tw=1)

1✔

31

        w = Parameter('w', dimensions=1, tw=1, init='init_constant', init_params={'value': 1})

1✔

32

        t = Parameter('t', dimensions=1, tw=1)

1✔

33

        c_v = Constant('c_v', tw=1, values=[[1], [2]])

1✔

34

        c = 5

1✔

35

        w_5 = Parameter('w_5', dimensions=1, tw=5)

1✔

36

        t_5 = Parameter('t_5', dimensions=1, tw=5)

1✔

37

        c_5 = [[1], [2], [3], [4], [5], [6], [7], [8], [9], [10]]

1✔

38

        c_5_2 = Constant('c_5_2', tw=5, values=c_5)

1✔

39

        parfun_x = ParamFun(myFun, parameters_and_constants=[K_x,c_v])

1✔

40

        parfun_y = ParamFun(myFun, parameters_and_constants=[K_y])

1✔

41

        parfun_zz = ParamFun(myFun)

1✔

42

        parfun_2d = ParamFun(myFun, parameters_and_constants=[K_x, K_x])

1✔

43

        parfun_3d = ParamFun(myFun, parameters_and_constants=[K_x])

1✔

44

        fir_w = Fir(W=w_5)(x.tw(5))

1✔

45

        fir_t = Fir(W=t_5)(y.tw(5))

1✔

46

        time_part = TimePart(x.tw(5), i=1, j=3)

1✔

47

        sample_select = SampleSelect(x.sw(5), i=1)

1✔

49

        def fuzzyfun(x):

1✔

52

        def fuzzyfunth(x):

1✔

55

        fuzzy = Fuzzify(output_dimension=4, range=[0, 4], functions=fuzzyfun)(x.tw(1))

1✔

56

        fuzzyTriang = Fuzzify(centers=[1, 2, 3, 7])(x.tw(1))

1✔

57

        fuzzyRect = Fuzzify(centers=[1, 2, 3, 7], functions='Rectangular')(x.tw(1))

1✔

58

        fuzzyList = Fuzzify(centers=[1, 3, 2, 7], functions=[fuzzyfun,fuzzyfunth])(x.tw(1))

1✔

60

        self.out = Output('out', Fir(parfun_x(x.tw(1)) + parfun_y(y.tw(1), c_v)))

1✔

61

        self.out2 = Output('out2', Add(w, x.tw(1)) + Add(t, y.tw(1)) + Add(w, c))

1✔

62

        self.out3 = Output('out3', Add(fir_w, fir_t))

1✔

63

        self.out4 = Output('out4', Linear(output_dimension=1)(fuzzy+fuzzyTriang+fuzzyRect+fuzzyList))

1✔

64

        self.out5 = Output('out5', Fir(time_part) + Fir(sample_select))

1✔

65

        self.out6 = Output('out6', LocalModel(output_function=Fir())(x.tw(1), fuzzy))

1✔

66

        self.out7 = Output('out7', parfun_zz(z.last()))

1✔

67

        self.out8 = Output('out8', Fir(parfun_x(x.tw(1)) + parfun_y(y.tw(1), c_v)) + Fir(parfun_zz(x.tw(5), t_5, c_5_2)))

1✔

68

        self.out9 = Output('out9', Fir(parfun_2d(x.tw(1)) + parfun_3d(x.tw(1),x.tw(1))))

1✔

70

    def test_export_textvisualizer(self):

1✔

71

        t = TextVisualizer(5)

1✔

72

        test = Modely(visualizer=t, seed=42, workspace='./results')

1✔

73

        test.addModel('modelA', self.out)

1✔

74

        test.addModel('modelB', [self.out2, self.out3, self.out4])

1✔

75

        test.addModel('modelC', [self.out4, self.out5, self.out6])

1✔

76

        test.addModel('modelD', self.out7)

1✔

77

        test.addMinimize('error1', self.x.last(), self.out)

1✔

78

        test.addMinimize('error2', self.y.last(), self.out3, loss_function='rmse')

1✔

79

        test.addMinimize('error3', self.z.last(), self.out6, loss_function='rmse')

1✔

81

        test.neuralizeModel(0.5)

1✔

83

        data_x = np.arange(0.0, 5, 0.1)

1✔

84

        data_y = np.arange(0.0, 5, 0.1)

1✔

85

        a, b = -1.0, 2.0

1✔

86

        dataset = {'x': data_x, 'y': data_y, 'z': a * data_x + b * data_y}

1✔

87

        params = {'num_of_epochs': 10, 'lr': 0.01}

1✔

88

        test.loadData(name='dataset', source=dataset)  # Create the datastest.trainModel(optimizer='SGD', training_params=params)  # Train the traced model

1✔

89

        t.showMinimize('error1')

1✔

90

        t.showMinimize('error2')

1✔

91

        t.showMinimize('error3')

1✔

92

        test.trainModel(optimizer='SGD', training_params=params)

1✔

93

        t.showWeights()

1✔

94

        test.trainModel(optimizer='SGD', training_params=params, closed_loop={'x':'out'}, prediction_samples=1)

1✔

95

        test.saveModel()

1✔

96

        test.loadModel()

1✔

98

        test.neuralizeModel(0.5)

1✔

99

        test.exportPythonModel()

1✔

100

        test.importPythonModel()

1✔

101

        test.exportReport()

1✔

103

    def test_export_mplvisualizer(self):

1✔

104

        m = MPLVisualizer(5)

1✔

105

        test = Modely(visualizer=m, seed=42)

1✔

106

        test.addModel('modelA', self.out)

1✔

107

        test.addModel('modelB', [self.out2, self.out3, self.out4])

1✔

108

        test.addModel('modelC', [self.out4, self.out5, self.out6])

1✔

109

        test.addModel('modelD', self.out7)

1✔

110

        test.addModel('modelE', self.out9)

1✔

111

        test.addMinimize('error1', self.x.last(), self.out)

1✔

112

        test.addMinimize('error2', self.y.last(), self.out3, loss_function='rmse')

1✔

113

        test.addMinimize('error3', self.z.last(), self.out6, loss_function='rmse')

1✔

115

        test.neuralizeModel(0.5)

1✔

117

        data_x = np.arange(0.0, 10, 0.1)

1✔

118

        data_y = np.arange(0.0, 10, 0.1)

1✔

119

        a, b = -1.0, 2.0

1✔

120

        dataset = {'x': data_x, 'y': data_y, 'z': a * data_x + b * data_y}

1✔

121

        params = {'num_of_epochs': 10, 'lr': 0.01}

1✔

122

        test.loadData(name='dataset', source=dataset)  # Create the dataset

1✔

123

        test.trainModel(optimizer='SGD', training_params=params)  # Train the traced model

1✔

124

        test.trainModel(optimizer='SGD', training_params=params)

1✔

125

        m.closeResult()

1✔

126

        m.closeTraining()

1✔

127

        list_of_functions = list(test.json['Functions'].keys())

1✔

128

        try:

1✔

129

            for f in list_of_functions:

1✔

130

                m.showFunctions(f)

1✔

131

        except ValueError:

1✔

132

            pass

1✔

133

        with self.assertRaises(ValueError):

1✔

134

            m.showFunctions(list_of_functions[1])

1✔

135

        m.closeFunctions()

1✔

137

    def test_export_mplvisualizer2(self):

1✔

138

        clearNames(['x', 'F'])

1✔

139

        x = Input('x')

1✔

140

        F = Input('F')

1✔

141

        def myFun(K1, K2, p1, p2):

1✔

142

            import torch

1✔

143

            return p1 * K1 + p2 * torch.sin(K2)

1✔

145

        parfun = ParamFun(myFun)

1✔

146

        out = Output('fun', parfun(x.last(), F.last()))

1✔

147

        m = MPLVisualizer()

1✔

148

        example = Modely(visualizer=m)

1✔

149

        example.addModel('out', out)

1✔

150

        example.neuralizeModel()

1✔

151

        m.showFunctions(list(example.json['Functions'].keys()), xlim=[[-5, 5], [-1, 1]])

1✔

152

        m.closeFunctions()

1✔

154

    def test_export_mplnotebookvisualizer(self):

1✔

155

        m = MPLNotebookVisualizer(5, test=True)

1✔

156

        test = Modely(visualizer=m, seed=42)

1✔

157

        test.addModel('modelB', [self.out2, self.out3, self.out4])

1✔

158

        test.addModel('modelC', [self.out4, self.out5, self.out6])

1✔

159

        test.addModel('modelD', [self.out9])

1✔

160

        test.addMinimize('error2', self.y.last(), self.out3, loss_function='rmse')

1✔

161

        test.addMinimize('error3', self.z.last(), self.out6, loss_function='rmse')

1✔

163

        test.neuralizeModel(1)

1✔

165

        data_x = np.arange(0.0, 10, 0.1)

1✔

166

        data_y = np.arange(0.0, 10, 0.1)

1✔

167

        a, b = -1.0, 2.0

1✔

168

        dataset = {'x': data_x, 'y': data_y, 'z': a * data_x + b * data_y}

1✔

169

        params = {'num_of_epochs': 1, 'lr': 0.01}

1✔

170

        test.loadData(name='dataset', source=dataset)  # Create the dataset

1✔

171

        test.trainModel(optimizer='SGD', training_params=params)  # Train the traced model

1✔

172

        list_of_functions = list(test.json['Functions'].keys())

1✔

173

        try:

1✔

174

            for f in list_of_functions:

1✔

175

                m.showFunctions(f)

1✔

tonegas / nnodely / 14359872492

Source File
Press 'n' to go to next uncovered line, 'b' for previous

1	import sys, os, unittest, torch, shutil	1✔
2	import numpy as np	1✔
3
4	from nnodely import *	1✔
5	from nnodely.basic.relation import NeuObj	1✔
6	from nnodely.support.logger import logging, nnLogger	1✔
7
8	log = nnLogger(__name__, logging.CRITICAL)	1✔
9	log.setAllLevel(logging.CRITICAL)	1✔
10
11	sys.path.append(os.getcwd())	1✔
12
13	# 3 Tests
14	# Test of visualizers
15
16	class ModelyTestVisualizer(unittest.TestCase):	1✔
17	def __init__(self, args, *kwargs):	1✔
18	NeuObj.clearNames()	1✔
19	super(ModelyTestVisualizer, self).__init__(args, *kwargs)	1✔
20
21	self.x = x = Input('x')	1✔
22	self.y = y = Input('y')	1✔
23	self.z = z = Input('z')	1✔
24
25	## create the relations
26	def myFun(K1, p1, p2):	1✔
27	return K1 * p1 * p2	1✔
28
29	K_x = Parameter('k_x', dimensions=1, tw=1, init='init_constant', init_params={'value': 1})	1✔
30	K_y = Parameter('k_y', dimensions=1, tw=1)	1✔
31	w = Parameter('w', dimensions=1, tw=1, init='init_constant', init_params={'value': 1})	1✔
32	t = Parameter('t', dimensions=1, tw=1)	1✔
33	c_v = Constant('c_v', tw=1, values=[[1], [2]])	1✔
34	c = 5	1✔
35	w_5 = Parameter('w_5', dimensions=1, tw=5)	1✔
36	t_5 = Parameter('t_5', dimensions=1, tw=5)	1✔
37	c_5 = [[1], [2], [3], [4], [5], [6], [7], [8], [9], [10]]	1✔
38	c_5_2 = Constant('c_5_2', tw=5, values=c_5)	1✔
39	parfun_x = ParamFun(myFun, parameters_and_constants=[K_x,c_v])	1✔
40	parfun_y = ParamFun(myFun, parameters_and_constants=[K_y])	1✔
41	parfun_zz = ParamFun(myFun)	1✔
42	parfun_2d = ParamFun(myFun, parameters_and_constants=[K_x, K_x])	1✔
43	parfun_3d = ParamFun(myFun, parameters_and_constants=[K_x])	1✔
44	fir_w = Fir(W=w_5)(x.tw(5))	1✔
45	fir_t = Fir(W=t_5)(y.tw(5))	1✔
46	time_part = TimePart(x.tw(5), i=1, j=3)	1✔
47	sample_select = SampleSelect(x.sw(5), i=1)	1✔
48
49	def fuzzyfun(x):	1✔
50	return torch.sin(x)	×
51
52	def fuzzyfunth(x):	1✔
53	return torch.tanh(x)	×
54
55	fuzzy = Fuzzify(output_dimension=4, range=[0, 4], functions=fuzzyfun)(x.tw(1))	1✔
56	fuzzyTriang = Fuzzify(centers=[1, 2, 3, 7])(x.tw(1))	1✔
57	fuzzyRect = Fuzzify(centers=[1, 2, 3, 7], functions='Rectangular')(x.tw(1))	1✔
58	fuzzyList = Fuzzify(centers=[1, 3, 2, 7], functions=[fuzzyfun,fuzzyfunth])(x.tw(1))	1✔
59
60	self.out = Output('out', Fir(parfun_x(x.tw(1)) + parfun_y(y.tw(1), c_v)))	1✔
61	self.out2 = Output('out2', Add(w, x.tw(1)) + Add(t, y.tw(1)) + Add(w, c))	1✔
62	self.out3 = Output('out3', Add(fir_w, fir_t))	1✔
63	self.out4 = Output('out4', Linear(output_dimension=1)(fuzzy+fuzzyTriang+fuzzyRect+fuzzyList))	1✔
64	self.out5 = Output('out5', Fir(time_part) + Fir(sample_select))	1✔
65	self.out6 = Output('out6', LocalModel(output_function=Fir())(x.tw(1), fuzzy))	1✔
66	self.out7 = Output('out7', parfun_zz(z.last()))	1✔
67	self.out8 = Output('out8', Fir(parfun_x(x.tw(1)) + parfun_y(y.tw(1), c_v)) + Fir(parfun_zz(x.tw(5), t_5, c_5_2)))	1✔
68	self.out9 = Output('out9', Fir(parfun_2d(x.tw(1)) + parfun_3d(x.tw(1),x.tw(1))))	1✔
69
70	def test_export_textvisualizer(self):	1✔
71	t = TextVisualizer(5)	1✔
72	test = Modely(visualizer=t, seed=42, workspace='./results')	1✔
73	test.addModel('modelA', self.out)	1✔
74	test.addModel('modelB', [self.out2, self.out3, self.out4])	1✔
75	test.addModel('modelC', [self.out4, self.out5, self.out6])	1✔
76	test.addModel('modelD', self.out7)	1✔
77	test.addMinimize('error1', self.x.last(), self.out)	1✔
78	test.addMinimize('error2', self.y.last(), self.out3, loss_function='rmse')	1✔
79	test.addMinimize('error3', self.z.last(), self.out6, loss_function='rmse')	1✔
80
81	test.neuralizeModel(0.5)	1✔
82
83	data_x = np.arange(0.0, 5, 0.1)	1✔
84	data_y = np.arange(0.0, 5, 0.1)	1✔
85	a, b = -1.0, 2.0	1✔
86	dataset = {'x': data_x, 'y': data_y, 'z': a * data_x + b * data_y}	1✔
87	params = {'num_of_epochs': 10, 'lr': 0.01}	1✔
88	test.loadData(name='dataset', source=dataset) # Create the datastest.trainModel(optimizer='SGD', training_params=params) # Train the traced model	1✔
89	t.showMinimize('error1')	1✔
90	t.showMinimize('error2')	1✔
91	t.showMinimize('error3')	1✔
92	test.trainModel(optimizer='SGD', training_params=params)	1✔
93	t.showWeights()	1✔
94	test.trainModel(optimizer='SGD', training_params=params, closed_loop={'x':'out'}, prediction_samples=1)	1✔
95	test.saveModel()	1✔
96	test.loadModel()	1✔
97
98	test.neuralizeModel(0.5)	1✔
99	test.exportPythonModel()	1✔
100	test.importPythonModel()	1✔
101	test.exportReport()	1✔
102
103	def test_export_mplvisualizer(self):	1✔
104	m = MPLVisualizer(5)	1✔
105	test = Modely(visualizer=m, seed=42)	1✔
106	test.addModel('modelA', self.out)	1✔
107	test.addModel('modelB', [self.out2, self.out3, self.out4])	1✔
108	test.addModel('modelC', [self.out4, self.out5, self.out6])	1✔
109	test.addModel('modelD', self.out7)	1✔
110	test.addModel('modelE', self.out9)	1✔
111	test.addMinimize('error1', self.x.last(), self.out)	1✔
112	test.addMinimize('error2', self.y.last(), self.out3, loss_function='rmse')	1✔
113	test.addMinimize('error3', self.z.last(), self.out6, loss_function='rmse')	1✔
114
115	test.neuralizeModel(0.5)	1✔
116
117	data_x = np.arange(0.0, 10, 0.1)	1✔
118	data_y = np.arange(0.0, 10, 0.1)	1✔
119	a, b = -1.0, 2.0	1✔
120	dataset = {'x': data_x, 'y': data_y, 'z': a * data_x + b * data_y}	1✔
121	params = {'num_of_epochs': 10, 'lr': 0.01}	1✔
122	test.loadData(name='dataset', source=dataset) # Create the dataset	1✔
123	test.trainModel(optimizer='SGD', training_params=params) # Train the traced model	1✔
124	test.trainModel(optimizer='SGD', training_params=params)	1✔
125	m.closeResult()	1✔
126	m.closeTraining()	1✔
127	list_of_functions = list(test.json['Functions'].keys())	1✔
128	try:	1✔
129	for f in list_of_functions:	1✔
130	m.showFunctions(f)	1✔
131	except ValueError:	1✔
132	pass	1✔
133	with self.assertRaises(ValueError):	1✔
134	m.showFunctions(list_of_functions[1])	1✔
135	m.closeFunctions()	1✔
136
137	def test_export_mplvisualizer2(self):	1✔
138	clearNames(['x', 'F'])	1✔
139	x = Input('x')	1✔
140	F = Input('F')	1✔
141	def myFun(K1, K2, p1, p2):	1✔
142	import torch	1✔
143	return p1 * K1 + p2 * torch.sin(K2)	1✔
144
145	parfun = ParamFun(myFun)	1✔
146	out = Output('fun', parfun(x.last(), F.last()))	1✔
147	m = MPLVisualizer()	1✔
148	example = Modely(visualizer=m)	1✔
149	example.addModel('out', out)	1✔
150	example.neuralizeModel()	1✔
151	m.showFunctions(list(example.json['Functions'].keys()), xlim=[[-5, 5], [-1, 1]])	1✔
152	m.closeFunctions()	1✔
153
154	def test_export_mplnotebookvisualizer(self):	1✔
155	m = MPLNotebookVisualizer(5, test=True)	1✔
156	test = Modely(visualizer=m, seed=42)	1✔
157	test.addModel('modelB', [self.out2, self.out3, self.out4])	1✔
158	test.addModel('modelC', [self.out4, self.out5, self.out6])	1✔
159	test.addModel('modelD', [self.out9])	1✔
160	test.addMinimize('error2', self.y.last(), self.out3, loss_function='rmse')	1✔
161	test.addMinimize('error3', self.z.last(), self.out6, loss_function='rmse')	1✔
162
163	test.neuralizeModel(1)	1✔
164
165	data_x = np.arange(0.0, 10, 0.1)	1✔
166	data_y = np.arange(0.0, 10, 0.1)	1✔
167	a, b = -1.0, 2.0	1✔
168	dataset = {'x': data_x, 'y': data_y, 'z': a * data_x + b * data_y}	1✔
169	params = {'num_of_epochs': 1, 'lr': 0.01}	1✔
170	test.loadData(name='dataset', source=dataset) # Create the dataset	1✔
171	test.trainModel(optimizer='SGD', training_params=params) # Train the traced model	1✔
172	list_of_functions = list(test.json['Functions'].keys())	1✔
173	try:	1✔
174	for f in list_of_functions:	1✔
175	m.showFunctions(f)	1✔
NEW 176	except ValueError:	×
NEW 177	pass	×

tonegas / nnodely / 14359872492

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous