Differences

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

Link to this comparison view

talksposters:spsearch [2024/04/29 10:16]
lspitler
talksposters:spsearch [2024/04/29 11:18] (current)
lspitler
Line 6: Line 6:
  
 [[https://github.com/scottransom/presto/blob/master/docs/PRESTO_search_tutorial.pdf | Link to PRESTO tutorial in PRESTO's github repo]] [[https://github.com/scottransom/presto/blob/master/docs/PRESTO_search_tutorial.pdf | Link to PRESTO tutorial in PRESTO's github repo]]
 +
 +Notes:   
 +  * I providing the commands outside of singularity. Run them with singularity as you prefer. 
 +  * I assume that the input data is in filterbank format and done my best to note where commands may differ in the case of PSRFITS format. 
  
 ===== What you need ===== ===== What you need =====
  
-  - Data: short observation of a test pulsar +  - Data: short observation of a test pulsar (MY_DATA.fil) 
-  - Software: singularity image containing ''PRESTO''+  - Software: either a working ''PRESTO'' install or a singularity image containing ''PRESTO''
  
-===== Inspect data =====+===== Simple SP pipeline ===== 
 + 
 +=== Step 1: inspect data ===
  
 Become familiar with the structure and contents of the data using ''waterfaller.py''.  Become familiar with the structure and contents of the data using ''waterfaller.py''. 
Line 18: Line 24:
 Example command: Example command:
  
-''waterfaller.py %%--%%show-ts %%--%%show-spec %%--%%colour-map='viridis' -d 57 -T 100 -t 5 %%--%%downsamp=8 %%--%%scaleindep DATA''+''waterfaller.py %%--%%show-ts %%--%%show-spec %%--%%colour-map='viridis' -d 57 -T 100 -t 5 %%--%%downsamp=8 %%--%%scaleindep MY_DATA.fil'' 
 + 
 +Parameters to play around with: 
 +  * Turn on and off dedispersion 
 +  * Go to a different time in the file 
 +  * Change the downsample factor 
 +  * Look at data with and without independently scaling each frequency channel 
 + 
 +=== Step 2: apply RFI mitigation === 
 + 
 +''PRESTOs'' removes RFI by calculating a mask. The mask informtion is saved in a number of files, one of which can be passed to the following steps. Therefore, a new, cleaned copy of the original data file is not produced.  
 + 
 +Example command: 
 + 
 +''> rfifind -o mask_file_name -time 1.0 -zapchan N:N MY_DATA.fil'' 
 + 
 +Parameters to play around with: 
 +  * Duration of processing time (-time) 
 +  * In the case of PSRFITS data: -noscales -nooffsetts 
 +  * Change thresholding parameters (-timesig, -freqsig, -chanfrac, -intfrac) 
 +  * Add channels to manually zap (-zapchan) 
 + 
 +Note, ''rfifind'' can be quite slow to run, but it is possible under some circumstances to recalculate the mask from the mask itself rather than rerunning the entire code. 
 + 
 +Data file: MY_DATA.fil \\ 
 +Original mask: mask_file_name_rfifind.mask 
 + 
 +''> rfifind -o mask_file_name -nocompute -freqsig 8 -mask mask_file_name_rfifind.mask MY_DATA.fil'' 
 + 
 +=== Step 3: generate dedispersed time series === 
 + 
 +This command assumes the data are for B0355+54 - change the DM range accordingly if using a different test pulsar.  
 + 
 +This command can be used to generate a single dedipsersed time series by setting -numdms 1. Having dedispered time series over a range around the true value is useful for in Step 4.  
 + 
 +''> prepsubband -nobary -lodm 52.4 -dmstep 0.5 -numdms 20 -clip 0 -downsamp 1 -mask mask_file_name_rfifind.mask -o my_ts MY_DATA.fil'' 
 + 
 +Variations: 
 +  * PSRFITS files: if you are searching a PSRFITS file and applied ''%%--%%noscales %%-nooffsets%%'' in the ''RFIfind'' stage, you should do so here too. 
 +  * Change the DM step and DM range (and see its impact on Step 4) 
 +  * Change the downsampling factor 
 +  * Run with clipping to see if this changes the time series 
 + 
 +Manually inspect the time series: 
 + 
 +''> exploredat my_ts.dat'' 
 + 
 +  * Can you see bright pulses? 
 +  * Note the slowly varying 
 +  * If you've applied an RFIfind mask, what impact does this have on the time series? 
 + 
 +=== Step 4: generate single pulse candidates === 
 + 
 +''> single_pulse_search.py -m 30 -t 6 -b *.dat'' 
 + 
 +Parameters to vary: 
 +  * Turn on bad block flagging 
 +  * Use different detrend chunksizes  
 + 
 +Look at SP diagnostic plot: my_ts_singlepulse.ps 
 + 
 +Recreate this plot with a different threshold or time range: 
 + 
 +''> single_pulse_search.py -m 30 -t 8 -s 0 -e 120 -b *.singlepulse'' 
 + 
 +=== Step 5: machine learning classifier ===
  
 +To be done...
  
 
talksposters/spsearch.1714378580.txt.gz ยท Last modified: 2024/04/29 10:16 by lspitler     Back to top