Using large .daq files

2011 3 4
2 个回答
回答已采纳
7 次查看（30 天）

0 个投票

How do I figure out how many data points are in a large .daq file if the file is too large to open at once? My files range from 2 GB to 28 GB. I can only read a portion of them at a time by using:
daqread('Myfile.daq', 'Samples', [1 10000000]);
I want to loop through the large file in increments but I don't know when to terminate the loop. Without knowing how many data points there are my loop will inevitably attempt to access a point that does not exist.

请先登录，再回答此问题。

 采纳的回答

Chris
Chris 2011-3-7

0 个投票

It turns out that daqread() will accept sample ranges greater than what exists in the file being read. So my solution to processing my data is the following:
DataChunkSize=1e7;
ChunksProcessed=0;
while DataChunkSize==1e7
Data=daqread('Myfile.daq', 'Samples',...
[DataChunkSize*ChunksProcessed+1,...
DataChunkSize*(ChunksProcessed+1)]);
DataChunkSize=numel(Data);
%%%%My data processing
ChunksProcessed=ChunksProcessed+1;
end
This works fine except the last loop will cause a warning to be spit out in the command window. It is slightly frustrating that the warning message tells you how many "samples" are in the .daq file, but there seems to be no way to get this information from Matlab. Oh well. At least the code can run to completion. Thanks for your help Walter.

更多回答（1 个）

Walter Roberson
Walter Roberson 2011-3-4

0 个投票

When I look at the documentation, I do not see any easy way to calculate or display the number of samples available. I would have expected daqread('Myfile.daq','info') to return information such as that, but it does not appear to.
What happens when you attempt to read a sample that does not exist?
If need be, you could put the daqread() in a try/catch block -- though narrowing in on the actual number of samples might possibly be a nuisance.

类别

在 帮助中心 和 File Exchange 中查找有关 File Operations 的更多信息

产品

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by