使用matlab绘制caffe的loss图像

发表于

之前见过有人使用shell画loss图像，我是在是学不来，于是自己动手写了一段matlab代码，把数据读到matlab里面。

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
fid=fopen('/data2/xxx/workspace/output.txt');
tline = fgetl(fid);
while ischar(tline)
    sub_idx_iteration=strfind(tline,'Iteration');
    if(sub_idx_iteration)
        sub_loss = strfind(tline,'loss');
        if(sub_loss)
        idx_end=strfind(tline,',');
        iteration = str2num(tline(sub_idx_iteration+10:idx_end-1));
        total_loss=str2num(tline(sub_loss+7:end));
        idx=iteration/5+1;
        matrix_for_plot(idx,1)=iteration;
        matrix_for_plot(idx,14)=total_loss;
        for i=1:12
            tline=fgetl(fid);
            stage_loss_start=strfind(tline,'* 1 =')+6;
            stage_loss=str2num(tline(stage_loss_start:end-5));
            matrix_for_plot(idx,i+1)=stage_loss;
        end
        end
    end
    tline = fgetl(fid);
end
figure(1)
plot(matrix_for_plot(:,1),matrix_for_plot(:,13));
ylim([0,200])
xlabel('Iteration')
ylabel('L2 (Joints)')
saveas(1,'L2.jpg');
figure(2)
plot(matrix_for_plot(:,1),matrix_for_plot(:,12));
xlabel('Iteration')
ylabel('L1 (PAF)')
saveas(2,'L1.jpg');
figure(3)
plot(matrix_for_plot(:,1),matrix_for_plot(:,14));
ylim([0,3000])
xlabel('Iteration')
ylabel('total loss')
saveas(3,'total_loss.jpg');

相关文章