-
Notifications
You must be signed in to change notification settings - Fork 1
/
find_critical_temp_run.py
141 lines (110 loc) · 4.95 KB
/
find_critical_temp_run.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
if __name__ != '__main__':
assert False, 'This file is not meant to be imported'
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('--filename', type=str, required=True) # data/hotrg_gilt_X24_Tc
parser.add_argument('--nLayers', type=int, required=True) # 60
parser.add_argument('--max_dim', type=int, required=True) # 24
parser.add_argument('--model', type=str, required=True) # 'Ising2D'
parser.add_argument('--param_name', type=str, required=True) # 'beta'
parser.add_argument('--param_min', type=float, required=True) # 0.43068679350977147
parser.add_argument('--param_max', type=float, required=True) # 0.45068679350977147
parser.add_argument('--observable_name', type=str, required=True) # 'magnetization'
parser.add_argument('--tol', type=float, default=1e-8)
parser.add_argument('--gilt_enabled', action='store_true')
parser.add_argument('--gilt_eps', type=float, default=8e-7)
parser.add_argument('--gilt_nIter', type=int, default=1)
parser.add_argument('--mcf_enabled', action='store_true')
parser.add_argument('--mcf_eps', type=float, default=1e-16)
parser.add_argument('--mcf_max_iter', type=int, default=200)
parser.add_argument('--hotrg_sanity_check', action='store_true')
parser.add_argument('--version', type=int, default=1)
parser.add_argument('--device', type=str, default='cuda:0')
args = parser.parse_args()
options=vars(args)
args = parser.parse_args()
options=vars(args)
print('loading library...')
from opt_einsum import contract # idk why but its required to avoid bug in contract with numpy arrays
import torch
import numpy as np
torch.set_default_tensor_type(torch.cuda.DoubleTensor)
if options['device']=='cpu':
torch.set_default_tensor_type(torch.DoubleTensor)
else:
torch.set_default_tensor_type(torch.cuda.DoubleTensor)
device=torch.device(options['device'])
torch.cuda.set_device(device)
from HOTRGZ2 import HOTRG_layers,trace_tensor,forward_observable_tensor,trace_two_tensors
from TNModels import Models
Model=Models[options['model']]
params=Model.get_default_params()
param_name=options['param_name']
def eval_model(params):
model=Model(params)
T0,(T0_op,checkerboard)=model.get_T0(),model.get_observables()[options['observable_name']]
layers,Ts,logTotals=HOTRG_layers(T0,
max_dim=options['max_dim'],nLayers=options['nLayers'],
options=options,return_tensors=True)
Ts,T_ops,logTotals=forward_observable_tensor(T0,T0_op,
layers=layers,checkerboard=checkerboard,
return_layers=True,cached_Ts=Ts)
T=Ts[-1]/Ts[-1].norm()
logZ=(trace_tensor(T).log()+logTotals[-1])/2**options['nLayers']
dNorm=torch.tensor([T.norm() for T in Ts]) # according to Lyu, this can be used to determine the phase transition
#dNorm=T.norm()
obs=trace_two_tensors(T_ops[-1])/trace_two_tensors(Ts[-1])
return T,logZ,obs,dNorm
beta_min=options['param_min']
beta_max=options['param_max']
beta_ref=Model.get_default_params()[param_name]
print('evaluating model at beta_min...')
params[param_name]=beta_min
T_min,logZ_min,obs_min,dNorm_min=eval_model(params)
print('evaluating model at beta_max...')
params[param_name]=beta_max
T_max,logZ_max,obs_max,dNorm_max=eval_model(params)
print('beta_min=',beta_min,'beta_max=',beta_max)
print('logZ_min=',logZ_min.item(),'logZ_max=',logZ_max.item())
print('obs_min=',obs_min.item(),'obs_max=',obs_max.item())
print('searching for critical temperature using bisection method')
beta_new=(beta_min+beta_max)/2
while beta_max-beta_min>options['tol']:
beta_new=(beta_min+beta_max)/2
if beta_new==beta_min or beta_new==beta_max:
break
params[param_name]=beta_new
T_new,logZ_new,obs_new,dNorm_new=eval_model(params)
print('beta_min=',beta_min,'beta_new=',beta_new,'beta_max=',beta_max,'beta_ref',beta_ref)
print('logZ_min=',logZ_min.item(),'logZ_new=',logZ_new.item(),'logZ_max=',logZ_max.item())
print('obs_min=',obs_min.item(),'obs_new=',obs_new.item(),'obs_max=',obs_max.item())
#dist_min=(logZ_min-logZ_new).abs()
#dist_max=(logZ_max-logZ_new).abs()
#dist_min=contract('ijkl,ijkl->',T_min,T_new).abs()
#dist_max=contract('ijkl,ijkl->',T_max,T_new).abs()
# dist_min=(obs_min-obs_new).abs()
# dist_max=(obs_max-obs_new).abs()
dist_min=(dNorm_min-dNorm_new).norm()
dist_max=(dNorm_max-dNorm_new).norm()
print('dist_min=',dist_min,'dist_max=',dist_max)
if dist_min<dist_max:
print('keeping beta_max')
beta_min=beta_new
T_min=T_new
logZ_min=logZ_new
obs_min=obs_new
dNorm_min=dNorm_new
else:
print('keeping beta_min')
beta_max=beta_new
T_max=T_new
logZ_max=logZ_new
obs_max=obs_new
dNorm_max=dNorm_new
print('critical temperature found: beta=',beta_new,' reference: ',beta_ref)
filename=options['filename']
import os
dirname=os.path.dirname(filename)
if not os.path.exists(dirname):
os.makedirs(dirname)
torch.save({param_name:beta_new},filename)