Hi Paddy,
From what I gather, you possess a timetable containing two columns: 'Timestamp' and 'Stat'. The 'Stat' column contains fault codes that range up to 100 unique values, though typically you only observe around 5 to 6 distinct codes in any given month.
Further, you're looking forward to calculate the monthly frequency for each fault codes in an efficient manner that avoids creating excessive bins for codes that don't appear.
As a workaround, you may use a data structure of nested maps, with the primary map keyed by year-month and the secondary map tracking the counts of fault codes for that specific year and month.
This approach has better space complexity but worse time complexity than your suggested solution.
Here's the code snippet for the above mentioned method for your understanding
% Initialize the map to hold the frequency data
faultFrequencyMap = containers.Map('KeyType', 'char', 'ValueType', 'any');
% Loop through each row of the timetable
for i = 1:height(tt)
% Extract the year and month as a string key (e.g., '2024-01')
yearMonthKey = datestr(tt.Timestamp(i), 'yyyy-mm');
% Check if this year-month key already exists in the map
if isKey(faultFrequencyMap, yearMonthKey)
% If the key exists, retrieve the existing fault code map
monthMap = faultFrequencyMap(yearMonthKey);
else
% If the key does not exist, create a new map for fault codes
monthMap = containers.Map('KeyType', 'int32', 'ValueType', 'int32');
faultFrequencyMap(yearMonthKey) = monthMap;
end
% Get the fault code for this row
faultCode = tt.Stat(i);
% Update the fault code frequency in the monthMap
if isKey(monthMap, faultCode)
% If the fault code exists, increment the count
monthMap(faultCode) = monthMap(faultCode) + 1;
else
% If the fault code does not exist, initialize the count to 1
monthMap(faultCode) = 1;
end
% Update the monthMap in the faultFrequencyMap
faultFrequencyMap(yearMonthKey) = monthMap;
end
% Display the faultFrequencyMap
keys = faultFrequencyMap.keys;
for k = 1:length(keys)
disp(['Year-Month: ', keys{k}]);
monthMap = faultFrequencyMap(keys{k});
faultCodes = monthMap.keys;
for f = 1:length(faultCodes)
disp([' Fault Code: ', num2str(faultCodes{f}), ...
', Frequency: ', num2str(monthMap(faultCodes{f}))]);
end
end
Hope this answers your query
Regards
Vinayak luha
