arrays - Python: comparing columns in numpy -

i writing program creates file add_to_file function , program takes file input.

the file has 4 columns: name, start position(s), description, end position(s)

what program supposed do, read file numpy array , create plausible "motion-sequences" creating array have @ least 1 of end positions in motion same @ least 1 of start positions in motion follow

def make_sequence(size,array):     count = 0     sequence = [array[0]]     result = [array[0,0]]     row in array[1:size]:         if row[1] == sequence[count][3]:             sequence.append(row)             result.append(row[0])             count += 1         else:             pass     return result 

what doing, it's taking array made file , making new array (result) such new array names of movements, movements next eachother flow because starting positions exact same ending positions of prior motion! here issue:

some movements can start , end in variety of positions, , have written can take 1 starting , 1 ending position per motion.

question: how make if string position values comma separates values ie(a,b,c,d), able iterate through them , compare them start end function return motions flow?

edit:

here example of want code able do

input:

row_1 = ["name1", "1,3,4,5", "description", "2"] row_2 = ["name2", "5,3", "description", "4"] row_3 = ["name3", "2", "description", "1"] row_4 = ["name4", "1", "description", "7,3"] row_5 = ["name5", "3", "description", "5,9,4"] row_6 = ["name6", "2", "description", "7"] row_7 = ["name7", "7", "description", "2"] row_8 = ["name8", "1", "description", "4"] row_9 = ["name9", "4", "description", "2"] data = np.array((column_index,row_1,row_2,row_3,row_4,row_5,row_6,row_7,row_8,row_9)) print make_sequence(7,array) 

output:

[name1,name3,name4,name5] 

you output because starts automatically row_1, has name: "name1", reads next row , if row has starting position same ending position previous row appended correct sequence, it's name appended result array.

and first 7 rows since input 7 size

well, there 2 questions here. 1 data processing: how extract required information raw data file. want names (column 0), start position (first element of column 1), , end positions (last element of column 3). might like:

import numpy np  dat = np.array([["name1", "1,3,4,5", "description", "2"],     ["name2", "5,3", "description", "4"],     ["name3", "2", "description", "1"],     ["name4", "1", "description", "7,3"],     ["name5", "3", "description", "5,9,4"],     ["name6", "2", "description", "7"],     ["name7", "7", "description", "2"],     ["name8", "1", "description", "4"],     ["name9", "4", "description", "2"]])  #since want 7 initial columns dat = dat[0:7,:]  #extract second , fourth columns c1 = dat[:,1] c2 = dat[:,3] names = dat[:,0]  #get start position via string manipulation startpos = [] item in c1:     startpos.append(item.split(',')[0])  startpos = np.array(startpos).astype(np.int32)  #get end position via string manipulation endpos = [] item in c2:     endpos.append(item.split(',')[-1])  endpos = np.array(endpos).astype(np.int32) print startpos print endpos 

cool. need follow startpos , endpos, so:

def followpath(a1,a2):     needle = 1     path = []     ii,item,newneedle in zip(np.arange(a1.size)+1,a1,a2):         if item == needle:             path.append(ii)             needle = newneedle     return np.array(path) - 1  thepath = followpath(startpos,endpos) print names[thepath] 

this gives me:

['name1' 'name3' 'name4' 'name5']