-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathmindist_correction_plots.m
166 lines (149 loc) · 6.07 KB
/
mindist_correction_plots.m
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
name = 'effect_of_mindist_on_curves_2022';
figure;
dbfile = 'decoding_with_mindist.db';
conn = sqlite(dbfile);
%carry out sql query using session_id to avoid future ambiguity
% sessions with mindist tests: DecodeTensor.qid([1 2 4 5 6 7])
mindist_set = 0:5:15;
m_i_indices = [1 2 4 5 6 7];
res = db_knife(conn, 'neuron_series', DecodeTensor.qid(m_i_indices),...
'settings_set', {'unshuffled', 'shuffled'},...
'mindist_set', mindist_set);
%%
my_lines = [];
for i = m_i_i:m_i_i
for j = 1:2
for k = [1 3 4]
setts = {'unshuffled', 'shuffled'};
my_lines = [my_lines DecodingPlotGenerator.errors_plotter(res.n{i,j,k}, res.m{i,j,k}, res.e{i,j,k}, setts{j}, 'index', k, 'DisplayName', sprintf('Dist. = %d px', mindist_set(k)))];
end
end
end
xlabel 'Number of cells'
ylabel '1/MSE (cm^{-2})'
xlim([0 500]);
text(300, 0.25/6, 'Unshuffled', 'Color', 'blue');
text(400, 0.6/6, 'Shuffled', 'Color', 'red');
legend(my_lines, 'Location', 'northwest');
%at the end
conn.close;
d_ = DecodeTensor(m_i_indices(m_i_i));
title(d_.mouse_name);
return;
%%
%todo: ball and stick plots showing no difference in curve params after
%fits
%changes to N?
figure;
subplot(2,2,1);
ballnstick('Unrestricted', '10 px minimal distance', res.N_fit(:, 1, 1), res.N_fit(:, 1, 3), res.N_fit_errb(:, 1, 1), res.N_fit_errb(:, 1, 3));
set(gca, 'YScale', 'log');
ylabel('N fit (neurons)');
title('Unshuffled');
%changes to I_0?
subplot(2,2,2);
ballnstick('Unrestricted', '10 px minimal distance', 1e4*res.I_0_fit(:, 1, 1), 1e4*res.I_0_fit(:, 1, 3), 1e4*res.I_0_fit_errb(:, 1, 1), 1e4*res.I_0_fit_errb(:, 1, 3));
%set(gca, 'YScale', 'log');
ylabel('I_0 fit (m^{-2}/neuron)');
title('Unshuffled');
%changes to N?
subplot(2,2,3);
ballnstick('Unrestricted', '10 px minimal distance', res.N_fit(:, 2, 1), res.N_fit(:, 2, 3), res.N_fit_errb(:, 2, 1), res.N_fit_errb(:, 2, 3));
set(gca, 'YScale', 'log');
ylabel('N fit (neurons)');
title('Shuffled');
%changes to I_0?
subplot(2,2,4);
ballnstick('Unrestricted', '10 px minimal distance', 1e4*res.I_0_fit(:, 2, 1), 1e4*res.I_0_fit(:, 2, 3), 1e4*res.I_0_fit_errb(:, 2, 1), 1e4*res.I_0_fit_errb(:, 2, 3));
%set(gca, 'YScale', 'log');
ylabel('I_0 fit (m^{-2}/neuron)');
title('Shuffled');
%function res = db_knife(conn, series_type, mice_session_set, settings_set, mindist_set, aggregate_flag, diff_flag)
function res = db_knife(varargin)
p = inputParser;
p.addRequired('conn', @(x)true);
p.addRequired('series_type', @ischar);
p.addRequired('mice_session_set', @(x) (ischar(x) || iscell(x)));
p.addParameter('settings_set', 'unshuffled', @(x) (ischar(x) || iscell(x)));
p.addParameter('mindist_set', 0, @isnumeric);
p.addParameter('aggregate_flag', false, @islogical);
p.addParameter('diff_flag', false, @islogical);
p.parse(varargin{:});
r = p.Results;
if strcmp(r.series_type, 'neuron_series')
primary_field = 'NumNeurons';
secondary_field = 'DataSize';
elseif strcmp(r.series_type, 'datasize_series')
primary_field = 'DataSize';
secondary_field = 'NumNeurons';
else
error('Only ''neuron_series'' or ''datasize_series'' is allowed as series_type');
end
basic_fetcher = @(sess, sett, mindist)...
sprintf('select %s, MSE from decoding where SessionID = ''%s'' and Setting = ''%s'' and MinDist = %f and %s = (select max(%s) from decoding where SessionID = ''%s'' and Setting = ''%s'' and MinDist = %f);',...
primary_field, sess, sett, mindist, secondary_field, secondary_field, sess, sett, mindist);
if ischar(r.mice_session_set)
sessions = {r.mice_session_set};
else
sessions = r.mice_session_set;
end
if ischar(r.settings_set)
settings = {settings_set};
else
settings = r.settings_set;
end
for i_s = 1:numel(sessions)
my_sess = sessions{i_s};
for j_set = 1:numel(settings)
my_setting = settings{j_set};
for k_md = 1:numel(r.mindist_set)
my_mindist = r.mindist_set(k_md);
[res.n{i_s, j_set, k_md}, res.m{i_s, j_set, k_md}, res.e{i_s, j_set, k_md},...
res.N_fit(i_s, j_set, k_md), res.I_0_fit(i_s, j_set, k_md),...
res.N_fit_errb(i_s, j_set, k_md), res.I_0_fit_errb(i_s, j_set, k_md)] =...
aggregate_one_curve(my_sess, my_setting, my_mindist);
end
%correcting fit bias, to use the same support for n values
n_upper_limit = min(cellfun(@max, res.n(i_s, j_set, :)));
for k_md = 1:numel(r.mindist_set)
my_mindist = r.mindist_set(k_md);
[~, ~, ~,...
res.N_fit(i_s, j_set, k_md), res.I_0_fit(i_s, j_set, k_md),...
res.N_fit_errb(i_s, j_set, k_md), res.I_0_fit_errb(i_s, j_set, k_md)] =...
aggregate_one_curve(my_sess, my_setting, my_mindist, n_upper_limit);
end
end
end
function [n, m, e, N_fit, I_0_fit, N_fit_errb, I_0_fit_errb] = aggregate_one_curve(sess, sett, mindist, n_upper_limit)%TODO fill out
if ~exist('n_upper_limit', 'var')
n_upper_limit = Inf;
else
fprintf('Using upper limit for n : %d\n', n_upper_limit);
end
db_ret = r.conn.fetch(basic_fetcher(sess, sett, mindist));
[n_rows, n_cols] = size(db_ret);
n_vals = cell2mat(db_ret(:,1));
imse_vals = 1./cell2mat(db_ret(:,2));
n = unique(n_vals);
for i = 1:numel(n)
my_n = n(i);
my_samples = imse_vals(n_vals == my_n);
m(i) = mean(my_samples);
e(i) = std(my_samples)./sqrt(length(my_samples))*norminv((1+0.95)/2);
end
filt_n = double(n(n <= n_upper_limit));
filt_m = m(n <= n_upper_limit);
[fitresult, gof] = createFit_infoSaturation(filt_n(:), filt_m(:));
N_fit = fitresult.N;
I_0_fit = fitresult.I_0;
conf_errbs = diff(confint(fitresult))/2;
I_0_fit_errb = conf_errbs(1);
N_fit_errb = conf_errbs(2);
end
function [n_agg, m_agg, e_agg] = aggregate_many_curves(n_cell, m_cell)%TODO fill out
end
end
%%knifelang:
% neuron_series [2022 2023 2024] {unshuffled shuffled}
% neuron_series [2022 2023 2024 2029] {unshuffled diagonal} -agg -diff
% neuron_series [2022] {unshuffled shuffled} <0 5 10 15 20 25>