# Differences

This shows you the differences between two versions of the page.

 faq:how_can_i_do_time-frequency_analysis_on_continuous_data [2014/09/18 08:27]robert faq:how_can_i_do_time-frequency_analysis_on_continuous_data [2015/01/30 16:31]134.76.103.230 [How can I do time-frequency analysis on continuous data?] 2015/01/30 16:31 [How can I do time-frequency analysis on continuous data?] 2014/09/19 15:10 [How can I do time-frequency analysis on continuous data?] 2014/09/18 08:27 robert 2014/09/18 08:26 robert created 2015/01/30 16:31 [How can I do time-frequency analysis on continuous data?] 2014/09/19 15:10 [How can I do time-frequency analysis on continuous data?] 2014/09/18 08:27 robert 2014/09/18 08:26 robert created Line 45: Line 45: The time or latency of each trial can be constructed using the sampleinfo from the segmented data, which specified for each trial the begin and the end-sample relative in the original datafile.  ​ The time or latency of each trial can be constructed using the sampleinfo from the segmented data, which specified for each trial the begin and the end-sample relative in the original datafile.  ​ - ​time = (data_segmented.sampleinfo(:,​1)+data_segmented.sampleinfo(:,​2))/​data_segmented.fsample;​ + ​begsample ​= data_segmented.sampleinfo(:,​1); + endsample = data_segmented.sampleinfo(:,​2); + time = ((begsample+endsample)/​2) / data_segmented.fsample;​ - Then we proceed by copying the freq structure, in which we flip the power spectrum to change the "rot" dimension into the "​time"​ dimension: ​ + Then we proceed by copying the freq structure, in which we flip the power spectrum to change the "rpt" dimension into the "​time"​ dimension: ​ freq_continuous ​          = freq_segmented; ​ freq_continuous ​          = freq_segmented; ​