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
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
|
# FOR PROBLEM 8.11
# author: sotech117
using Statistics
using Plots
function wrap_index(i::Int, l::Int)::Int
wrap = (i - 1) % l + 1
return (wrap <= 0) ? l + wrap : wrap
end
mutable struct Ising2D
l::Int
n::Int
temperature::Float64
w::Vector{Float64} # Boltzmann weights
state::Matrix
energy::Int
magnetization::Int
mc_steps::Int
accepted_moves::Int
energy_array::Vector{Int}
magnetization_array::Vector{Int}
end
Ising2D(l::Int, temperature::Float64) = begin
n = l^2
w = zeros(9)
w[9] = exp(-8.0 / temperature)
w[5] = exp(-4.0 / temperature)
state = ones(Int, l, l) # initially all spins up
energy = -2 * n
magnetization = n
return Ising2D(l, n, temperature, w, state, energy, magnetization, 0, 0,
Int[], Int[])
end
function reset!(ising::Ising2D)
ising.mc_steps = 0
ising.accepted_moves = 0
ising.energy_array = Int[]
ising.magnetization_array = Int[]
end
function mc_step!(ising::Ising2D)
l::Int = ising.l
n::Int = ising.n
w = ising.w
state = ising.state
accepted_moves = ising.accepted_moves
energy = ising.energy
magnetization = ising.magnetization
random_positions = l * rand(2 * n)
random_array = rand(n)
for k in 1:n
i = trunc(Int, random_positions[2 * k - 1]) + 1
j = trunc(Int, random_positions[2 * k]) + 1
de = 2 * state[i, j] * (state[i % l + 1, j] +
state[wrap_index(i - 1, l), j] + state[i, j % l + 1] +
state[i, wrap_index(j - 1, l)])
if de <= 0 || w[de + 1] > random_array[k]
accepted_moves += 1
new_spin = - state[i, j] # flip spin
state[i, j] = new_spin
energy += de
magnetization += 2 * new_spin
end
end
ising.state = state
ising.accepted_moves = accepted_moves
ising.energy = energy
ising.magnetization = magnetization
append!(ising.energy_array, ising.energy)
append!(ising.magnetization_array, ising.magnetization)
ising.mc_steps = ising.mc_steps + 1
end
function steps!(ising::Ising2D, num::Int=100)
for i in 1:num
mc_step!(ising)
end
end
function mean_energy(ising::Ising2D)
return mean(ising.energy_array) / ising.n
end
function specific_heat(ising::Ising2D)
return (std(ising.energy_array) / ising.temperature) ^ 2 / ising.n
end
function mean_magnetization(ising::Ising2D)
return mean(ising.magnetization_array) / ising.n
end
function susceptibility(ising::Ising2D)
return (std(ising.magnetization_array)) ^ 2 / (ising.temperature * ising.n)
end
function observables(ising::Ising2D)
printstyled("Temperature\t\t", bold=true)
print(ising.temperature); print("\n")
printstyled("Mean Energy\t\t", bold=true)
print(mean_energy(ising)); print("\n")
printstyled("Mean Magnetiz.\t\t", bold=true)
print(mean_magnetization(ising)); print("\n")
printstyled("Specific Heat\t\t", bold=true)
print(specific_heat(ising)); print("\n")
printstyled("Susceptibility\t\t", bold=true)
print(susceptibility(ising)); print("\n")
printstyled("MC Steps\t\t", bold=true)
print(ising.mc_steps); print("\n")
printstyled("Accepted Moves\t\t", bold=true)
print(ising.accepted_moves); print("\n")
end
function plot_ising(state::Matrix{Int})
pos = Tuple.(findall(>(0), state))
neg = Tuple.(findall(<(0), state))
scatter(pos, markersize=5)
scatter!(neg, markersize=5)
end
function get_magnetization(T, n=1000)
m = Ising2D(64, T)
steps!(m, n)
println("done with T = $T")
return mean_magnetization(m)
end
Ts = 0:.1:5
ms = [abs(get_magnetization(T)) for T in Ts]
println("done with calculating magnetizations")
function linear_regression(x, y)
n = length(x)
x̄ = sum(x) / n
ȳ = sum(y) / n
a = sum((x .- x̄) .* (y .- ȳ)) / sum((x .- x̄).^2)
b = ȳ - a * x̄
return (a, b)
end
# plot M^(8) over T
betas = .001:.001:1
residuals = []
for i in 1:length(betas)
b = betas[i]
m = ms .^ (1 / b)
# filter out the zero values
s = scatter(p,
Ts, m, xlabel="T (units of J / k_b)", ylabel="Magnetization", label="$b-beta", title="Magnetization vs Temp (Ising Monte Carlo)",
msw=0, ms=1.5, mc=:red, lc=:red, lw=1.5, legend=:bottomleft
)
# do a linear regression
a, b = linear_regression(Ts, m)
# plot a linear regression line
plot!(s, Ts, a*Ts .+ b, label="Linear Regression", lw=1.2, color=:red, linestyle=:dash)
# calculate the residuals
push!(residuals, sum((m .- (a*Ts .+ b)).^2))
savefig(s, "hw6/b/8-2-$i.png")
end
# plot the residuals over beta
plot(betas, residuals, xlabel="beta", ylabel="Squared Distance", label="Residuals", title="Error from Linear Regression of M^(1/Beta)", lw=1.2, lc=:red, legend=:topright)
# find the min on the first half of the residuals
min_residuali = argmin(residuals[1:div(length(residuals), 2)])
min_residual = betas[min_residuali]
println("Minimum Residual: ", min_residual)
vline!([min_residual], label="Minimum Point @ Beta = $min_residual", lc=:orange, lw=1.5, ls=:dash)
# plot the analityical beta of .125
vline!([.125], label="Analytical Minimum Beta = .125", lc=:green, lw=1.5, ls=:dot)
savefig("hw6/8-2-residuals-100.png")
|