def test_classe_threetip_threestate(self):
root = self.threetiptree
data = self.threetipdata
f = sse.make_classe(root,
data,
nstate=3,
condition_on_surv=False,
pi="Equal")
params = np.array([
0.3, 0.1, 0.7, 0.01, 0.8, 0.05, 0.01, 0.1, 0.9, 0.2, 0.004, 0.3,
0.3, 0.11, 0.14, 0.15, 0.17, 0.19, 0.01, 0.01, 0.1, 0.2, 0.4, 0.1,
0.01, 0.7, 0.6
])
calc_lik_ncos = f(params)
true_lik_ncos = -9.6298
f = sse.make_classe(root,
data,
nstate=3,
condition_on_surv=True,
pi="Equal")
calc_lik_cos = f(params)
true_lik_cos = -10.00689
self.assertTrue(np.isclose(calc_lik_ncos, true_lik_ncos, atol=1e-3))
self.assertTrue(np.isclose(calc_lik_cos, true_lik_cos, atol=1e-3))
def test_classe_20tip2state(self):
root = self.twentytiptree
data = self.twentytipdata
params = np.array(
[0.3, 0.1, 0.01, 0.01, 0.1, 0.2, 0.01, 0.01, 0.2, 0.1])
f = sse.make_classe(root, data, 2, None, False, pi="Equal")
calc_lik_ncos = f(params)
true_lik_ncos = -86.35204
self.assertTrue(np.isclose(calc_lik_ncos, true_lik_ncos, atol=1e-2))
f = sse.make_classe(root, data, 2, None, True, pi="Equal")
calc_lik_cos = f(params)
true_lik_cos = -85.28472
self.assertTrue(np.isclose(calc_lik_cos, true_lik_cos, atol=1e-2))
def test_classe_100tip_2state(self):
root = self.hundredtiptree
data = self.hundredtipdata
params = np.array(
[0.3, 0.1, 0.01, 0.01, 0.1, 0.2, 0.01, 0.01, 0.2, 0.1])
f = sse.make_classe(root, data, 2, None, False, pi="Equal")
calc_lik_ncos = f(params)
true_lik_ncos = -452.8292
f = sse.make_classe(root, data, 2, None, True, pi="Equal")
calc_lik_cos = f(params)
print(calc_lik_cos)
true_lik_cos = -451.7423
self.assertTrue(np.isclose(calc_lik_ncos, true_lik_ncos, atol=1e-1))
self.assertTrue(np.isclose(calc_lik_cos, true_lik_cos, atol=1e-1))
def test_classe_threetip_twostate(self):
root = self.threetiptree
data = self.threetipdata
lambdaidx = np.array([[[1, 2], [0, 3]], [[4, 5], [0, 6]]])
muidx = np.array([7, 8])
qidx = np.array([[0, 9], [10, 0]])
pidx = {"lambda": lambdaidx, "mu": muidx, "q": qidx}
f = sse.make_classe(root, data, 2, pidx, False, pi="Equal")
params = np.array(
[0.3, 0.1, 0.01, 0.01, 0.1, 0.2, 0.01, 0.01, 0.2, 0.1])
calc_lik_ncos = f(params)
true_lik_ncos = -5.962547
f = sse.make_classe(root, data, 2, pidx, True, pi="Equal")
params = np.array(
[0.3, 0.1, 0.01, 0.01, 0.1, 0.2, 0.01, 0.01, 0.2, 0.1])
calc_lik_cos = f(params)
true_lik_cos = -4.912378
self.assertTrue(np.isclose(calc_lik_ncos, true_lik_ncos, atol=1e-3))
self.assertTrue(np.isclose(calc_lik_cos, true_lik_cos, atol=1e-3))
def test_classe_100tip_geosseequiv(self):
root = self.hundredtiptree
data = self.hundredtipdata
lambdaidx = np.array([[[0, 1, 2], [0, 0, 3], [0, 0, 0]],
[[0, 0, 0], [0, 1, 0], [0, 0, 0]],
[[0, 0, 0], [0, 0, 0], [0, 0, 2]]])
muidx = np.array([0, 4, 5])
qidx = np.array([[0, 5, 4], [6, 0, 0], [7, 0, 0]])
pidx = {"lambda": lambdaidx, "mu": muidx, "q": qidx}
params = np.array([
0.157769502155153, 0.157769502155153, 0.157769502155153,
0.0788847510775767, 0.0788847510775767, 0.0788847510775767,
0.0788847510775767
])
f = sse.make_classe(root, data, nstate=3, pidx=pidx)
calc_lik = f(params)
true_lik = -440.9617
self.assertTrue(np.isclose(calc_lik, true_lik, atol=1e-1))
from __future__ import absolute_import, division, print_function, unicode_literals
import ivy
from ivy.chars import mk_mr
import numpy as np
from ivy.chars import bayesian_models
import pymc
import matplotlib.pyplot as plt
from ivy.interactive import *
from ivy.vis import layers
from ivy.chars.sse import sse
root = ivy.tree.read("(((((sp57:4.044339169,(sp58:3.890662993,sp59:3.890662993)nd118:0.1536761762)nd111:1.417401845,((((sp75:3.172705141,(sp125:0.5527391639,sp126:0.5527391639)nd128:2.619965977)nd126:0.1491783771,sp71:3.321883518)nd121:0.1716653213,(sp111:1.329916304,sp112:1.329916304)nd122:2.163632536)nd119:0.5173899544,(sp62:3.482885335,sp63:3.482885335)nd120:0.5280534591)nd112:1.45080222)nd28:16.04986342,(((sp52:4.711305107,sp53:4.711305107)nd98:2.079117401,sp44:6.790422508)nd58:5.683724767,sp21:12.47414727)nd14:9.03745716)nd10:13.84492974,(((((sp81:2.783684725,(sp96:1.931056014,sp97:1.931056014)nd132:0.8526287114)nd55:9.9533422,sp18:12.73702692)nd38:1.512996403,((((((sp64:3.470751646,sp65:3.470751646)nd113:1.344122386,sp50:4.814874033)nd89:3.976689789,(sp69:3.328192332,sp70:3.328192332)nd90:5.46337149)nd88:0.02754570427,sp31:8.819109526)nd80:0.653847391,(sp79:2.817908157,(sp129:0.370641198,sp130:0.370641198)nd130:2.447266959)nd116:6.65504876)nd53:3.303754679,((sp38:9.117749002,(sp133:0.2778588591,sp134:0.2778588591)nd83:8.839890143)nd77:3.012840068,((sp137:0.07400931118,sp138:0.07400931118)nd74:10.10434534,(sp36:6.895546513,sp105:6.895546513)nd75:3.282808135)nd60:1.952234422)nd54:0.6461225269)nd39:1.473311731)nd26:4.269769285,(sp17:14.65578684,(sp54:4.704808503,((sp84:2.568626404,sp85:2.568626404)nd123:0.8723411983,(((sp101:1.714268281,sp102:1.714268281)nd137:0.02759929337,sp100:1.741867575)nd135:0.643384511,sp88:2.385252086)nd124:1.055715516)nd115:1.263840901)nd56:9.950978337)nd27:3.864005773)nd25:3.392164376,(sp14:15.11915802,((sp42:5.948314005,(sp98:1.843935789,sp99:1.843935789)nd109:4.104378217)nd66:4.996946603,(sp66:7.141750109,(sp48:5.383149397,sp49:5.383149397)nd93:1.758600712)nd67:3.803510499)nd35:4.173897408)nd31:6.792798972)nd8:13.44457719)nd5:10.70015921,(((((sp45:5.958477059,(sp107:1.503900566,sp108:1.503900566)nd108:4.454576494)nd64:5.427227553,(sp86:2.522861611,sp87:2.522861611)nd65:8.862843002)nd63:5.960631133,sp10:17.34633575)nd24:2.32921451,(((sp103:1.63925749,sp104:1.63925749)nd61:12.27804272,(((sp68:3.331939432,((sp94:1.994874638,sp95:1.994874638)nd131:0.7907986681,sp80:2.785673306)nd127:0.5462661258)nd110:9.657761967,(sp20:12.61435609,((sp115:1.093273853,sp116:1.093273853)nd102:5.49578194,sp37:6.589055793)nd57:6.025300299)nd52:0.3753453071)nd46:0.78118111,(sp60:3.853622154,sp61:3.853622154)nd48:9.917260354)nd43:0.1464177026)nd22:5.428430873,(((sp113:1.329796843,sp114:1.329796843)nd94:8.61369875,sp27:9.943495594)nd70:4.226538594,(((sp92:2.005629918,sp93:2.005629918)nd68:8.934427814,(((sp123:0.7745931871,sp124:0.7745931871)nd84:8.202637625,(sp89:2.308927308,(sp131:0.3602063283,sp132:0.3602063283)nd136:1.94872098)nd85:6.668303504)nd71:1.471438122,sp26:10.44866893)nd69:0.4913887978)nd49:2.389357191,((sp76:3.070054683,sp77:3.070054683)nd105:3.747331878,(((sp127:0.5455466061,sp128:0.5455466061)nd106:5.417574288,sp41:5.963120894)nd100:0.7261380398,(sp117:1.071580803,sp118:1.071580803)nd101:5.61767813)nd97:0.1281276275)nd50:6.512028362)nd41:0.8406192645)nd23:5.175696897)nd19:0.3298191716)nd16:0.5359322381,((((sp90:2.193767491,sp91:2.193767491)nd78:7.555963912,(sp121:0.85277711,sp122:0.85277711)nd79:8.896954293)nd32:6.289306951,((((sp109:1.37117621,sp110:1.37117621)nd133:1.33986203,(sp135:0.1976309559,sp136:0.1976309559)nd134:2.513407283)nd129:0.162954131,sp78:2.87399237)nd103:3.38367749,(sp72:3.248274386,sp73:3.248274386)nd104:3.009395474)nd33:9.781368494)nd20:3.487554657,(sp56:10.38265213,(sp119:0.9775309416,sp120:0.9775309416)nd114:9.405121191)nd21:9.143940879)nd17:0.6848894819)nd6:25.84521089)nd1;")
data = {'sp10':0,'sp14':0,'sp17':0,'sp18':0,'sp20':1,'sp21':1,'sp26':0,'sp27':0,'sp31':0,'sp36':0,'sp37':0,'sp38':1,'sp41':1,'sp42':0,'sp44':0,'sp45':1,'sp48':0,'sp49':0,'sp50':0,'sp52':1,'sp53':0,'sp54':1,'sp56':0,'sp57':0,'sp58':1,'sp59':1,'sp60':1,'sp61':0,'sp62':1,'sp63':1,'sp64':1,'sp65':0,'sp66':0,'sp68':1,'sp69':0,'sp70':1,'sp71':0,'sp72':0,'sp73':0,'sp75':0,'sp76':0,'sp77':0,'sp78':0,'sp79':1,'sp80':1,'sp81':0,'sp84':0,'sp85':0,'sp86':0,'sp87':0,'sp88':1,'sp89':0,'sp90':1,'sp91':1,'sp92':0,'sp93':0,'sp94':1,'sp95':1,'sp96':1,'sp97':1,'sp98':1,'sp99':0,'sp100':1,'sp101':1,'sp102':1,'sp103':1,'sp104':1,'sp105':1,'sp107':0,'sp108':1,'sp109':1,'sp110':1,'sp111':0,'sp112':0,'sp113':0,'sp114':0,'sp115':0,'sp116':0,'sp117':1,'sp118':1,'sp119':1,'sp120':0,'sp121':0,'sp122':0,'sp123':0,'sp124':0,'sp125':1,'sp126':1,'sp127':0,'sp128':0,'sp129':0,'sp130':0,'sp131':0,'sp132':0,'sp133':1,'sp134':1,'sp135':1,'sp136':1,'sp137':1,'sp138':1}
params = np.array([0.3,0.1,0.01,0.01,0.1,0.2,0.01,0.01,0.2,0.1])
import line_profiler
f = sse.make_classe(root,data,2,True,pi="Equal")
%timeit f(params)
profile = line_profiler.LineProfiler(f,sse.classe_node_calculation)
profile.runcall(f,params)
profile.print_stats()