-
Notifications
You must be signed in to change notification settings - Fork 0
/
TO DO
24 lines (18 loc) · 1.48 KB
/
TO DO
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
autocorrelation_experiment("first_autocorrelation_test_L=5", 5, 0.0, 10.0, 0.1, 50, collect(LinRange(3.0,0.9,10)), 10000, 10, 100000; verbose_annealing=true, original_configuration=[[2 3 3 2 6; 5 3 6 4 4; 1 2 5 1 5; 2 2 6 6 2; 2 1 2 3 2],[1 2 5 5 3; 6 6 1 2 3; 2 5 5 4 3; 4 1 6 5 4; 4 6 6 2 4], [3 4 5 4 6; 2 5 4 4 1; 5 3 2 5 4; 4 6 1 2 5; 2 5 5 4 6], [3 3 1 6 3; 2 3 1 6 3; 6 6 4 3 2; 1 1 2 3 2; 6 3 1 5 3], [2 3 4 4 1; 3 3 1 3 6; 3 4 4 5 1; 4 5 5 4 1; 6 1 1 6 6], [4 1 2 3 5; 2 1 6 6 5; 4 1 5 5 1; 2 1 5 1 6; 5 4 4 3 6]])
# Soon:
- Implement recursive ALL neighbour energy function for swap moves too - needed for manual exponential states growth with energy calculation
- Implement turning off parity swap moves
- Add order parameter measurements to relaxed_anneal and history_anneal
- Add saved error logs to relaxed_anneal and history_anneal
# -----
# Later:
- Clean up acceptable_neighbours_graph_plotter code to be less duplicated
- Ask ChatGPT to make code from GitHub more efficient. Use ProfileView again too.
- Optimise p_parity_exchange_swap for speed and effecacy, also ratio of 3-cycles to orientation rotations
- Add ensemble processes
# Maybe:
- Find consistently best relaxation_iterations_vector
- Replace number_of_three_cycle_swap_moves function with mathematical calulation
- Modify sample and all neighbour energy functions if want to swtich off coupled subystem two cycles there too
# For Paper:
- Type up 'Rubik's Cube Monte Carlo Supplementary Document' neatly to accomany codebase