In the test_bin_statistics test (currently disabled) there is a failure caused by comparisons of 0 to NaN. The underlying JSON data has NaN where bins are empty and the bin_statistics function returns zeros:
|
function test_bin_statistics(testCase) |
|
assumeFail(testCase, "Temporarily Disabled - Need to ask @hivanov about nan vs zero comparison") |
|
|
|
% create array containg wind speeds to use as bin edges |
|
bin_edges = 3:1:25; |
|
relative_file_name = '../../examples/data/loads/loads_data_dict.json'; |
|
full_file_name = fullfile(fileparts(mfilename('fullpath')), relative_file_name); |
|
fid = fopen(full_file_name); % Opening the file |
|
raw = fread(fid,inf); % Reading the contents |
|
str = char(raw'); % Transformation |
|
fclose(fid); % Closing the file |
|
data = jsondecode(str); % Using the jsondecode function to parse JSON from string |
|
|
|
% Apply function to calculate means |
|
load_means_struct = [data.means.uWind_80m]; |
|
load_means_table = struct2table(data.means); |
|
ta = struct2table([data.bin_means]); |
|
st = table2struct(ta,'ToScalar',true); |
|
ta_std = struct2table([data.bin_means_std]); |
|
st_std = table2struct(ta_std,'ToScalar',true); |
|
bin_against = load_means_struct;% load_means_table.uWind_80m; |
|
datast = bin_statistics(load_means_table,bin_against,bin_edges); |
|
assertEqual(testCase,st,datast.averages,'RelTol',0.001) |
|
assertEqual(testCase,st_std,datast.std,'RelTol',0.001) |
|
end |
@hivanov-nrel do you have any suggestions on how to fix? Should empty bins return NaN or 0. Do you think this is a bug in the test or in the bin_statistics function?
