1 | """ |
---|
2 | Unit tests for fitting module |
---|
3 | @author Gervaise Alina |
---|
4 | """ |
---|
5 | import unittest |
---|
6 | import math |
---|
7 | |
---|
8 | from sas.fit.AbstractFitEngine import Model, FitHandler |
---|
9 | from sas.dataloader.loader import Loader |
---|
10 | from sas.fit.Fitting import Fit |
---|
11 | from sas.models.LineModel import LineModel |
---|
12 | from sas.models.Constant import Constant |
---|
13 | |
---|
14 | class testFitModule(unittest.TestCase): |
---|
15 | """ test fitting """ |
---|
16 | |
---|
17 | def test_bad_pars(self): |
---|
18 | fitter = Fit('bumps') |
---|
19 | |
---|
20 | data = Loader().load("testdata_line.txt") |
---|
21 | data.name = data.filename |
---|
22 | fitter.set_data(data,1) |
---|
23 | |
---|
24 | model1 = LineModel() |
---|
25 | model1.name = "M1" |
---|
26 | model = Model(model1, data) |
---|
27 | pars1= ['param1','param2'] |
---|
28 | try: |
---|
29 | fitter.set_model(model,1,pars1) |
---|
30 | except ValueError,exc: |
---|
31 | #print "ValueError was correctly raised: "+str(msg) |
---|
32 | assert str(exc).startswith('parameter param1') |
---|
33 | else: |
---|
34 | raise AssertionError("No error raised for fitting with wrong parameters name to fit") |
---|
35 | |
---|
36 | def fit_single(self, fitter_name, isdream=False): |
---|
37 | fitter = Fit(fitter_name) |
---|
38 | |
---|
39 | data = Loader().load("testdata_line.txt") |
---|
40 | data.name = data.filename |
---|
41 | fitter.set_data(data,1) |
---|
42 | |
---|
43 | # Receives the type of model for the fitting |
---|
44 | model1 = LineModel() |
---|
45 | model1.name = "M1" |
---|
46 | model = Model(model1,data) |
---|
47 | |
---|
48 | pars1= ['A','B'] |
---|
49 | fitter.set_model(model,1,pars1) |
---|
50 | fitter.select_problem_for_fit(id=1,value=1) |
---|
51 | result1, = fitter.fit(handler=FitHandler()) |
---|
52 | |
---|
53 | # The target values were generated from the following statements |
---|
54 | p,s,fx = result1.pvec, result1.stderr, result1.fitness |
---|
55 | #print "p0,p1,s0,s1,fx = %g, %g, %g, %g, %g"%(p[0],p[1],s[0],s[1],fx) |
---|
56 | p0,p1,s0,s1,fx_ = 3.68353, 2.61004, 0.336186, 0.105244, 1.20189 |
---|
57 | |
---|
58 | if isdream: |
---|
59 | # Dream is not a minimizer: just check that the fit is within |
---|
60 | # uncertainty |
---|
61 | self.assertTrue( abs(p[0]-p0) <= s0 ) |
---|
62 | self.assertTrue( abs(p[1]-p1) <= s1 ) |
---|
63 | else: |
---|
64 | self.assertTrue( abs(p[0]-p0) <= 1e-5 ) |
---|
65 | self.assertTrue( abs(p[1]-p1) <= 1e-5 ) |
---|
66 | self.assertTrue( abs(fx-fx_) <= 1e-5 ) |
---|
67 | |
---|
68 | def fit_bumps(self, alg, **opts): |
---|
69 | #Importing the Fit module |
---|
70 | from bumps import fitters |
---|
71 | fitters.FIT_DEFAULT = alg |
---|
72 | fitters.FIT_OPTIONS[alg].options.update(opts) |
---|
73 | fitters.FIT_OPTIONS[alg].options.update(monitors=[]) |
---|
74 | #print "fitting",alg,opts |
---|
75 | #kprint "options",fitters.FIT_OPTIONS[alg].__dict__ |
---|
76 | self.fit_single('bumps', isdream=(alg=='dream')) |
---|
77 | |
---|
78 | def test_bumps_de(self): |
---|
79 | self.fit_bumps('de') |
---|
80 | |
---|
81 | def test_bumps_dream(self): |
---|
82 | self.fit_bumps('dream', burn=500, steps=100) |
---|
83 | |
---|
84 | def test_bumps_amoeba(self): |
---|
85 | self.fit_bumps('amoeba') |
---|
86 | |
---|
87 | def test_bumps_newton(self): |
---|
88 | self.fit_bumps('newton') |
---|
89 | |
---|
90 | def test_bumps_lm(self): |
---|
91 | self.fit_bumps('lm') |
---|
92 | |
---|
93 | def test2(self): |
---|
94 | """ fit 2 data and 2 model with no constrainst""" |
---|
95 | #load data |
---|
96 | l = Loader() |
---|
97 | data1=l.load("testdata_line.txt") |
---|
98 | data1.name = data1.filename |
---|
99 | |
---|
100 | data2=l.load("testdata_line1.txt") |
---|
101 | data2.name = data2.filename |
---|
102 | |
---|
103 | #Importing the Fit module |
---|
104 | fitter = Fit('bumps') |
---|
105 | # Receives the type of model for the fitting |
---|
106 | model11 = LineModel() |
---|
107 | model11.name= "M1" |
---|
108 | model22 = LineModel() |
---|
109 | model11.name= "M2" |
---|
110 | |
---|
111 | model1 = Model(model11,data1) |
---|
112 | model2 = Model(model22,data2) |
---|
113 | pars1= ['A','B'] |
---|
114 | fitter.set_data(data1,1) |
---|
115 | fitter.set_model(model1,1,pars1) |
---|
116 | fitter.select_problem_for_fit(id=1,value=0) |
---|
117 | fitter.set_data(data2,2) |
---|
118 | fitter.set_model(model2,2,pars1) |
---|
119 | fitter.select_problem_for_fit(id=2,value=0) |
---|
120 | |
---|
121 | try: result1, = fitter.fit(handler=FitHandler()) |
---|
122 | except RuntimeError,msg: |
---|
123 | assert str(msg)=="Nothing to fit" |
---|
124 | else: raise AssertionError,"No error raised for fitting with no model" |
---|
125 | fitter.select_problem_for_fit(id=1,value=1) |
---|
126 | fitter.select_problem_for_fit(id=2,value=1) |
---|
127 | R1,R2 = fitter.fit(handler=FitHandler()) |
---|
128 | |
---|
129 | self.assertTrue( math.fabs(R1.pvec[0]-4)/3 <= R1.stderr[0] ) |
---|
130 | self.assertTrue( math.fabs(R1.pvec[1]-2.5)/3 <= R1.stderr[1] ) |
---|
131 | self.assertTrue( R1.fitness/(len(data1.x)+len(data2.x)) < 2) |
---|
132 | |
---|
133 | |
---|
134 | def test_constraints(self): |
---|
135 | """ fit 2 data and 2 model with 1 constrainst""" |
---|
136 | #load data |
---|
137 | l = Loader() |
---|
138 | data1= l.load("testdata_line.txt") |
---|
139 | data1.name = data1.filename |
---|
140 | data2= l.load("testdata_cst.txt") |
---|
141 | data2.name = data2.filename |
---|
142 | |
---|
143 | # Receives the type of model for the fitting |
---|
144 | model11 = LineModel() |
---|
145 | model11.name= "line" |
---|
146 | model11.setParam("A", 1.0) |
---|
147 | model11.setParam("B",1.0) |
---|
148 | |
---|
149 | model22 = Constant() |
---|
150 | model22.name= "cst" |
---|
151 | model22.setParam("value", 1.0) |
---|
152 | |
---|
153 | model1 = Model(model11,data1) |
---|
154 | model2 = Model(model22,data2) |
---|
155 | model1.set(A=4) |
---|
156 | model1.set(B=3) |
---|
157 | # Constraint the constant value to be equal to parameter B (the real value is 2.5) |
---|
158 | #model2.set(value='line.B') |
---|
159 | pars1= ['A','B'] |
---|
160 | pars2= ['value'] |
---|
161 | |
---|
162 | #Importing the Fit module |
---|
163 | fitter = Fit('bumps') |
---|
164 | fitter.set_data(data1,1) |
---|
165 | fitter.set_model(model1,1,pars1) |
---|
166 | fitter.set_data(data2,2,smearer=None) |
---|
167 | fitter.set_model(model2,2,pars2,constraints=[("value","line.B")]) |
---|
168 | fitter.select_problem_for_fit(id=1,value=1) |
---|
169 | fitter.select_problem_for_fit(id=2,value=1) |
---|
170 | |
---|
171 | R1,R2 = fitter.fit(handler=FitHandler()) |
---|
172 | self.assertTrue( math.fabs(R1.pvec[0]-4.0)/3. <= R1.stderr[0]) |
---|
173 | self.assertTrue( math.fabs(R1.pvec[1]-2.5)/3. <= R1.stderr[1]) |
---|
174 | self.assertTrue( R1.fitness/(len(data1.x)+len(data2.x)) < 2) |
---|
175 | |
---|
176 | |
---|
177 | def test4(self): |
---|
178 | """ fit 2 data concatenates with limited range of x and one model """ |
---|
179 | #load data |
---|
180 | l = Loader() |
---|
181 | data1 = l.load("testdata_line.txt") |
---|
182 | data1.name = data1.filename |
---|
183 | data2 = l.load("testdata_line1.txt") |
---|
184 | data2.name = data2.filename |
---|
185 | |
---|
186 | # Receives the type of model for the fitting |
---|
187 | model1 = LineModel() |
---|
188 | model1.name= "M1" |
---|
189 | model1.setParam("A", 1.0) |
---|
190 | model1.setParam("B",1.0) |
---|
191 | model = Model(model1,data1) |
---|
192 | |
---|
193 | pars1= ['A','B'] |
---|
194 | #Importing the Fit module |
---|
195 | |
---|
196 | fitter = Fit('bumps') |
---|
197 | fitter.set_data(data1,1,qmin=0, qmax=7) |
---|
198 | fitter.set_model(model,1,pars1) |
---|
199 | fitter.set_data(data2,1,qmin=1,qmax=10) |
---|
200 | fitter.select_problem_for_fit(id=1,value=1) |
---|
201 | result2, = fitter.fit(handler=FitHandler()) |
---|
202 | |
---|
203 | self.assert_(result2) |
---|
204 | self.assertTrue( math.fabs(result2.pvec[0]-4)/3 <= result2.stderr[0] ) |
---|
205 | self.assertTrue( math.fabs(result2.pvec[1]-2.5)/3 <= result2.stderr[1] ) |
---|
206 | self.assertTrue( result2.fitness/len(data1.x) < 2) |
---|
207 | |
---|
208 | |
---|
209 | if __name__ == "__main__": |
---|
210 | unittest.main() |
---|
211 | |
---|