Index: doc/theses/mike_brooks_MMath/plots/common.py =================================================================== --- doc/theses/mike_brooks_MMath/plots/common.py (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/common.py (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -17,12 +17,15 @@ dt[['corpus-basename', - 'corpus-ext']] = dt['corpus'].str.strip().str.split('.', expand=True) + 'corpus-ext']] = dt['corpus'].str.strip().str.rsplit('.', n=1, expand=True) dt[['corpus-slug', - 'corpus-nstrs', - 'corpus-meanlen', + 'corpus-nstrs-tgt', + 'corpus-meanlen-tgt', 'corpus-runid']] = dt['corpus-basename'].str.strip().str.split('-', expand=True) - dt["corpus-nstrs"] = pd.to_numeric(dt["corpus-nstrs"]) - dt["corpus-meanlen"] = pd.to_numeric(dt["corpus-meanlen"]) - dt["corpus-runid"] = pd.to_numeric(dt["corpus-runid"]) + dt["corpus-nstrs-tgt"] = pd.to_numeric(dt["corpus-nstrs-tgt"]) + dt["corpus-meanlen-tgt"] = pd.to_numeric(dt["corpus-meanlen-tgt"]) + dt[['corpus-relscale', + 'corpus-seed', + 'corpus-offset-instr']] = dt['corpus-runid'].str.strip().str.split('+', expand=True) + dt["corpus-relscale"] = pd.to_numeric(dt["corpus-relscale"]) def loadParseTimingData( infileLocal, xClasNames=[], xClasDtypes={}, xFactNames=[], xFactDtypes={} ): @@ -32,11 +35,11 @@ timings = pd.read_csv( infile, - names=['test', 'corpus'] + xClasNames + [ 'concatsPerReset', 'corpusItemCount', 'corpusMeanLenChars', 'concatDoneActualCount', 'execTimeActualSec'] + xFactNames, + names=['test', 'corpus'] + xClasNames + [ 'concatsPerReset', 'corpusItemCountAct', 'corpusMeanLenCharsAct', 'concatDoneActualCount', 'execTimeActualSec'] + xFactNames, dtype={**xClasDtypes, **xFactDtypes, **{ 'test': str, 'corpus': str, 'concatsPerReset': 'Int64', # allows missing; https://stackoverflow.com/a/70626154 - 'corpusItemCount': np.int64, - 'corpusMeanLenChars': np.float64, + 'corpusItemCountAct': np.int64, + 'corpusMeanLenCharsAct': np.float64, 'concatDoneActualCount': np.int64, 'execTimeActualSec': np.float64, Index: doc/theses/mike_brooks_MMath/plots/string-allocn-attrib-meta.dat =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-allocn-attrib-meta.dat (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) +++ doc/theses/mike_brooks_MMath/plots/string-allocn-attrib-meta.dat (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -0,0 +1,1 @@ +harness-leaf other ctor-dtor gc malloc-free text-import Index: doc/theses/mike_brooks_MMath/plots/string-allocn-attrib.py =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-allocn-attrib.py (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-allocn-attrib.py (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -14,4 +14,5 @@ import os import sys +from pathlib import Path sys.path.insert(0, os.path.dirname(__file__)) @@ -19,12 +20,20 @@ sizes_i_want = [50, 200] # [20, 50, 100, 200] +defaultExpansionCfa = 0.2 -# assume CFA threshold only run at default value +metaFilename = os.path.dirname(os.path.abspath(__file__)) + '/string-allocn-attrib-meta.dat' +catOrder = Path(metaFilename).read_text() +catOrder = str.split(catOrder) cfatimings = loadParseTimingData('result-allocate-speed-cfa.csv', xClasNames=['expansion'], xClasDtypes={'expansion':'Float64'}, xFactNames=['topIters'], xFactDtypes={'topIters':np.int64}) +cfatimings = cfatimings.groupby('expansion').get_group(defaultExpansionCfa) cfaattribs = loadParseAttribData('result-allocate-attrib-cfa.ssv') +cfaattribs = cfaattribs.groupby('expansion').get_group(defaultExpansionCfa) + +cfasizings = loadParseSizingData('result-allocate-space-cfa.ssv', xClasNames=['expansion'], xClasDtypes={'expansion':'Float64'}) +cfasizings = cfasizings.groupby('expansion').get_group(defaultExpansionCfa) stltimings = loadParseTimingData('result-allocate-speed-stl.csv', @@ -34,23 +43,91 @@ stlattribs = loadParseAttribData('result-allocate-attrib-stl.ssv') +stlsizings = loadParseSizingData('result-allocate-space-stl.ssv', xClasNames=['expansion'], xClasDtypes={'expansion':'Float64'}) + timings = pd.concat([cfatimings, stltimings]) attribs = pd.concat([cfaattribs, stlattribs]) +sizings = pd.concat([cfasizings, stlsizings]) + +# print("before join", timings.shape[0], attribs.shape[0]) +# print(timings.to_csv(header=True, index=True, sep='\t', na_rep="0")) +# print(attribs.to_csv(header=True, index=True, sep='\t', na_rep="0")) +# print(sizings.to_csv(header=True, index=True, sep='\t', na_rep="0")) combined = pd.merge( - left=timings[['sut-platform', 'corpus-meanlen','expansion', 'op-duration-ns']], - right=attribs[['sut-platform', 'corpus-meanlen','expansion', 'category', 'fraction']], - on=['sut-platform', 'corpus-meanlen','expansion'] + left=timings[['sut-platform', 'corpus', 'corpus-meanlen-tgt','expansion', 'op-duration-ns']], + right=attribs[['sut-platform', 'corpus', 'expansion', 'category', 'fraction']], + on=['sut-platform', 'corpus','expansion'] ) +combined = pd.merge( + left=combined, + right=sizings[['sut-platform', 'corpus','expansion','hw_cur_req_mem(B)']], + on=['sut-platform', 'corpus','expansion'] +) + +# print("after join", combined.shape[0]) +# print(combined.to_csv(header=True, index=True, sep='\t', na_rep="0")) + combined['cat-duration-ns'] = combined['op-duration-ns'] * combined['fraction'] -combined.drop(columns=['expansion', 'op-duration-ns', 'fraction'], inplace=True) +combined.drop(columns=['expansion', 'op-duration-ns', 'fraction', 'corpus'], inplace=True) -pvt = combined.pivot( columns='category', values='cat-duration-ns', index=['corpus-meanlen', 'sut-platform'] ) +# print("before summarize", combined.shape[0]) +# print(combined.to_csv(header=True, index=True, sep='\t', na_rep="0")) -desired_dcol_order = ["ctor-dtor", "gc", "malloc-free", "text-import", "harness-leaf", "other"] -pvt = pvt[desired_dcol_order] +summary = combined.pivot_table( + values=['hw_cur_req_mem(B)','cat-duration-ns'], + index=['corpus-meanlen-tgt', 'sut-platform', 'category'], + aggfunc={'hw_cur_req_mem(B)':'mean','cat-duration-ns':['mean', 'min', 'max']} ) +summary = summary.reset_index() +summary.columns = summary.columns.to_flat_index() +summary.columns = [ + '-'.join(filter(None, col)).replace(' ', '-') # replaces space with dash if needed + for col in summary.columns.to_flat_index() +] -filtered = pvt.loc[pvt.index.get_level_values('corpus-meanlen').isin(sizes_i_want)] +# reorder columns with memory as first value (after sut-platform, which is last key) +# cols = summary.columns.tolist() +# cols.remove("hw_cur_req_mem(B)-mean") +# insert_after = cols.index("sut-platform") + 1 +# cols.insert(insert_after, "hw_cur_req_mem(B)-mean") +# summary = summary[cols] -print(filtered.to_csv(header=True, index=True, sep='\t', na_rep="0")) +summary = summary[[ + 'corpus-meanlen-tgt', + 'sut-platform', + 'hw_cur_req_mem(B)-mean', + 'category', + 'cat-duration-ns-mean', + 'cat-duration-ns-max', + 'cat-duration-ns-min']] +# print("after summarize", summary.shape[0]) +# print(summary.to_csv(header=True, index=True, sep='\t', na_rep="0")) + + + +# Ensure 'category' follows the specified order +summary['category'] = pd.Categorical(summary['category'], categories=catOrder, ordered=True) + +# Sort the DataFrame to prepare for cumulative sum +summary_sorted = summary.sort_values(by=['corpus-meanlen-tgt', 'sut-platform', 'category']) + +# Group by the keys and compute exclusive running total +summary_sorted['grp-prior-duration-ns'] = ( + summary_sorted + .groupby(['corpus-meanlen-tgt', 'sut-platform'])['cat-duration-ns-mean'] + .transform(lambda s: s.cumsum().shift(fill_value=0)) +) + +summary_sorted = summary_sorted.reset_index(drop=True) + + +# print("after accumulation", summary_sorted.shape[0]) +# print(summary_sorted.to_csv(header=True, index=True, sep='\t', na_rep="0")) + + +filtered = summary_sorted[summary_sorted['corpus-meanlen-tgt'].isin(sizes_i_want)] + +# print("after filter", filtered.shape[0]) + +print(filtered.to_csv(header=True, index=False, sep='\t', na_rep="0")) Index: doc/theses/mike_brooks_MMath/plots/string-allocn.gp =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-allocn.gp (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-allocn.gp (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -5,4 +5,5 @@ INDIR="build" OUTDIR="build" +SRCDIR="plots" SCALE=1024 @@ -11,13 +12,7 @@ set macros set output OUTDIR."/plot-string-allocn.pdf" -set multiplot layout 1, 3 ; -set key outside top center horizontal +set multiplot layout 1, 2 ; set grid -# common to only first two graphs -set logscale x 2 -#set mxtics 3 # 3 steps within each doubling (e.g. 3 steps of of 32 between 32 and 128 => tick on 32s in there) -set xlabel "Heap Used (B)" -set logscale y 10 # @@ -25,9 +20,16 @@ # +#set errorbars dashtype '.' + +set logscale x 2 +set xlabel "Heap Used (B)" +set logscale y 10 +set key outside top center + set ylabel "Duration (ns)" offset 2,0 -set yrange[35:125] -set ytics add (40, 50, 60, 70, 80, 90, 110, 120) +set yrange[40:150] +set ytics add (40, 50, 60, 70, 80, 90, 110, 120, 130, 140, 150) -set xrange[32:4096] +set xrange[32:8192] set xtics rotate by -90 set xtics ("" 32, "64 k" 64, "" 128, "256 k" 256, "" 512, "1 M" 1024, "" 2048, "4 M" 4096) @@ -35,16 +37,39 @@ # First each curve, then each default-expansion point plot INDIR."/plot-string-allocn.dat" \ - i 0 using ($3/SCALE):2 title columnheader(1) with linespoints lt rgb "blue" pt 2 ps 1 lw 1, \ - '' i 1 using ($3/SCALE):2 title columnheader(1) with linespoints lt rgb "red" pt 3 ps 1 lw 1, \ - '' i 2 using ($3/SCALE):2 title columnheader(1) with linespoints lt rgb "brown" pt 8 ps 1 lw 1, \ - '' i 3 using ($3/SCALE):2 title columnheader(1) with linespoints lt rgb "black" pt 10 ps 1 lw 1, \ - '' i 4 using ($3/SCALE):2 title columnheader(1) with linespoints lt rgb "magenta" pt 12 ps 1 lw 1, \ - '' i 0 using ( ($4 == 1) ? ($3/SCALE) : 1/0 ):2 notitle with points lt rgb "blue" pt 66 ps 2, \ - '' i 1 using ( ($4 == 1) ? ($3/SCALE) : 1/0 ):2 notitle with points lt rgb "red" pt 66 ps 2, \ - '' i 2 using ( ($4 == 1) ? ($3/SCALE) : 1/0 ):2 notitle with points lt rgb "brown" pt 66 ps 2, \ - '' i 3 using ( ($4 == 1) ? ($3/SCALE) : 1/0 ):2 notitle with points lt rgb "black" pt 66 ps 2, \ - '' i 4 using ( ($4 == 1) ? ($3/SCALE) : 1/0 ):2 notitle with points lt rgb "magenta" pt 66 ps 2 + i 0 using ($5/SCALE):2 notitle with lines lt rgb "blue" dt '.' lw 2, \ + '' i 1 using ($5/SCALE):2 notitle with lines lt rgb "red" dt '.' lw 2, \ + '' i 2 using ($5/SCALE):2 notitle with lines lt rgb "brown" dt '.' lw 2, \ + '' i 3 using ($5/SCALE):2 title "tuning contour" with lines lt rgb "black" dt '.' lw 2, \ + '' i 4 using ($5/SCALE):2 notitle with lines lt rgb "magenta" dt '.' lw 2, \ + '' i 0 using ($5/SCALE):2:3:4 notitle with yerrorbars lt rgb "blue" pt 7 ps 0.25 lw 1, \ + '' i 1 using ($5/SCALE):2:3:4 notitle with yerrorbars lt rgb "red" pt 7 ps 0.25 lw 1, \ + '' i 2 using ($5/SCALE):2:3:4 notitle with yerrorbars lt rgb "brown" pt 7 ps 0.25 lw 1, \ + '' i 3 using ($5/SCALE):2:3:4 notitle with yerrorbars lt rgb "black" pt 7 ps 0.25 lw 1, \ + '' i 4 using ($5/SCALE):2:3:4 notitle with yerrorbars lt rgb "magenta" pt 7 ps 0.25 lw 1, \ + '' i 0 using ( ($8 == 1) ? ($5/SCALE) : 1/0 ):2 notitle with points lt rgb "blue" pt 66 ps 2, \ + '' i 1 using ( ($8 == 1) ? ($5/SCALE) : 1/0 ):2 notitle with points lt rgb "red" pt 66 ps 2, \ + '' i 2 using ( ($8 == 1) ? ($5/SCALE) : 1/0 ):2 notitle with points lt rgb "brown" pt 66 ps 2, \ + '' i 3 using ( ($8 == 1) ? ($5/SCALE) : 1/0 ):2 title "default tuning" \ + with points lt rgb "black" pt 66 ps 2, \ + '' i 4 using ( ($8 == 1) ? ($5/SCALE) : 1/0 ):2 notitle with points lt rgb "magenta" pt 66 ps 2, \ + '' i 0 using ($5/SCALE):2:("{/Times=15 len-\n20") every ::1::1 with labels left offset 0.75,0.25 textcolor rgb "blue" notitle, \ + '' i 1 using ($5/SCALE):2:("{/Times=15 len-\n50") every ::5::5 with labels right offset -0.6 ,1.5 textcolor rgb "red" notitle, \ + '' i 2 using ($5/SCALE):2:("{/Times=15 len-\n100") every ::4::4 with labels right offset -0.75,0.5 textcolor rgb "brown" notitle, \ + '' i 3 using ($5/SCALE):2:("{/Times=15 len-\n200") every ::4::4 with labels right offset -0.75,1.25 textcolor rgb "black" notitle, \ + '' i 4 using ($5/SCALE):2:("{/Times=15 len-\n500") every ::4::4 with labels right offset -0.75,0.25 textcolor rgb "magenta" notitle, \ + '' i 5 using ($5/SCALE) :2 notitle with lines lt rgb "#BF007F7F" lw 8, \ + '' i 6 using ($5/SCALE) :2 notitle with lines lt rgb "#BF007F7F" lw 8, \ + '' i 7 using ($5/SCALE) :2 notitle with lines lt rgb "#BF007F7F" lw 8, \ + '' i 8 using ($5/SCALE) :2 title "tradeoff" with lines lt rgb "#BF007F7F" lw 8, \ + '' i 9 using ($5/SCALE) :2 notitle with lines lt rgb "#BF007F7F" lw 8, \ + '' i 5 using ( (strcol(8) eq "stl") ? ($5/SCALE) : 1/0 ):2:3:4 notitle with yerrorbars lt rgb "#00007F7F" pt 6 ps 0.5 lw 2 , \ + '' i 6 using ( (strcol(8) eq "stl") ? ($5/SCALE) : 1/0 ):2:3:4 notitle with yerrorbars lt rgb "#00007F7F" pt 6 ps 0.5 lw 2 , \ + '' i 7 using ( (strcol(8) eq "stl") ? ($5/SCALE) : 1/0 ):2:3:4 notitle with yerrorbars lt rgb "#00007F7F" pt 6 ps 0.5 lw 2 , \ + '' i 8 using ( (strcol(8) eq "stl") ? ($5/SCALE) : 1/0 ):2:3:4 notitle with yerrorbars lt rgb "#00007F7F" pt 6 ps 0.5 lw 2 , \ + '' i 9 using ( (strcol(8) eq "stl") ? ($5/SCALE) : 1/0 ):2:3:4 notitle with yerrorbars lt rgb "#00007F7F" pt 6 ps 0.5 lw 2 + unset ylabel +unset xlabel unset xtics @@ -53,41 +78,4 @@ unset xrange - -# -# STL comparison -# - - -set yrange[40:85] -set ytics add (40, 45, 50, 55, 60, 65, 70, 75, 80, 85) -set mytics 90 # 90 steps within each decade (e.g. 90 steps of of 1 between 10 and 100 => tick on 1s in there) - -set xrange[64:4096] -set xtics rotate by -90 -set xtics ("64 k" 64, "128 k" 128, "256 k" 256, "512 k" 512, "1 M" 1024, "2 M" 2048, "4 M" 4096) - -# skullduggeries: -# hardcoding chunk index and assuming data in the chunk (by hardcoding only the stl series title) -# series order is meaningless but important: achieves z-order readability and legend order acceptability - -plot INDIR."/plot-string-allocn.dat" \ - i 8 using ($3/SCALE) :2 title "tradeoff" with lines lt rgb "#77000000" dt (2,2) lw 8, \ - '' i 1 using ($3/SCALE) :2 title columnheader(1) with linespoints lt rgb "red" pt 3 ps 1 lw 1, \ - '' i 6 using ( (strcol(4) eq "cfa") ? ($3/SCALE) : 1/0 ):2 notitle with points lt rgb "red" pt 66 ps 2, \ - '' i 6 using ($3/SCALE) :2 notitle with lines lt rgb "#77000000" dt (2,2) lw 8, \ - '' i 6 using ( (strcol(4) eq "stl") ? ($3/SCALE) : 1/0 ):2 title "stl, len=50" with points lt rgb "red" pt 5 ps 1, \ - '' i 3 using ($3/SCALE) :2 title columnheader(1) with linespoints lt rgb "black" pt 10 ps 1 lw 1, \ - '' i 8 using ( (strcol(4) eq "stl") ? ($3/SCALE) : 1/0 ):2 title "stl, len=200" with points lt rgb "black" pt 5 ps 1, \ - '' i 8 using ( (strcol(4) eq "cfa") ? ($3/SCALE) : 1/0 ):2 notitle with points lt rgb "black" pt 66 ps 2 - - -unset mytics -unset ytics -unset yrange -unset xrange - - - -# common to first two graphs unset logscale unset xlabel @@ -98,11 +86,56 @@ # -set style data histogram +set key outside top center vertical maxrows 3 + +set logscale x 2 +#set mxtics 3 # 3 steps within each doubling (e.g. 3 steps of of 32 between 32 and 128 => tick on 32s in there) + +set yrange[0:105] set ytics auto -set style histogram clustered gap 1 rowstacked -set style fill solid border -1 -set boxwidth 0.8 -set xtics rotate by -45 -plot for [col=3:8] \ - INDIR.'/plot-string-allocn-attrib.dat' using col:xticlabels(stringcolumn(2).", len=".stringcolumn(1)) index 0 title columnheader(col) +set xrange[2.0e4:3e7] +# set xtics rotate by -90 +# set xtics ("" 32, "64 k" 64, "" 128, "256 k" 256, "" 512, "1 M" 1024, "" 2048, "4 M" 4096) + +set style fill solid 1.0 + +BAR_RELWD = 0.5 +BAR_XSHIFT = (1+BAR_RELWD)**(-0.55) +LEN_X_SEP = 1.7 + +# get segment draw order (front to back) from metadata file +meta = system('cat '.SRCDIR.'/string-allocn-attrib-meta.dat') +STATS_words = words(meta) +array STATS_word[STATS_words] +do for [i=1:STATS_words] { + STATS_word[i] = word(meta, i) +} + + + +plot \ + for [i=STATS_words:1:-1] \ + INDIR.'/plot-string-allocn-attrib.dat' \ + using (stringcolumn("category") eq STATS_word[i] ? (BAR_XSHIFT*$1**LEN_X_SEP*$3/SCALE) : 1/0): \ + (stringcolumn("category") eq STATS_word[i] ? ($5 + $8) : 1/0): \ + (stringcolumn("category") eq STATS_word[i] ? (BAR_RELWD*(BAR_XSHIFT*$1**LEN_X_SEP*$3/SCALE)) : 1/0) \ + with boxes ls i title STATS_word[i], \ + INDIR."/plot-string-allocn.dat" \ + i 1 using (50 **LEN_X_SEP*$5/SCALE):2 notitle with lines lt rgb "red" dt '.' lw 2, \ + '' i 3 using (200**LEN_X_SEP*$5/SCALE):2 notitle with lines lt rgb "black" dt '.' lw 2, \ + '' i 1 using ( ($8 == 1) ? (50 **LEN_X_SEP*$5/SCALE) : 1/0 ):2 notitle with points lt rgb "red" pt 66 ps 2, \ + '' i 3 using ( ($8 == 1) ? (200**LEN_X_SEP*$5/SCALE) : 1/0 ):2 notitle with points lt rgb "black" pt 66 ps 2, \ + '' i 1 using ( ($8 == 1) ? (50 **LEN_X_SEP*$5/SCALE) : 1/0 ):2:3:4 notitle with yerrorbars lt rgb "red" pt 7 ps 0.25 lw 1, \ + '' i 3 using ( ($8 == 1) ? (200**LEN_X_SEP*$5/SCALE) : 1/0 ):2:3:4 notitle with yerrorbars lt rgb "black" pt 7 ps 0.25 lw 1, \ + '' i 1 using (50 **LEN_X_SEP*$5/SCALE):2:("{/Times=15 len-\n50") every ::5::5 with labels right offset -0.6 ,1.5 textcolor rgb "red" notitle, \ + '' i 3 using (200**LEN_X_SEP*$5/SCALE):2:("{/Times=15 len-\n200") every ::4::4 with labels right offset -0.75,1.25 textcolor rgb "black" notitle, \ + '' i 6 using (50 **LEN_X_SEP*$5/SCALE) :2 notitle with lines lt rgb "#BF007F7F" lw 8, \ + '' i 8 using (200**LEN_X_SEP*$5/SCALE) :2 notitle with lines lt rgb "#BF007F7F" lw 8, \ + '' i 6 using ( (strcol(8) eq "stl") ? (50 **LEN_X_SEP*$5/SCALE) : 1/0 ):2:3:4 notitle with yerrorbars lt rgb "#00007F7F" pt 6 ps 0.5 lw 2 , \ + '' i 8 using ( (strcol(8) eq "stl") ? (200**LEN_X_SEP*$5/SCALE) : 1/0 ):2:3:4 notitle with yerrorbars lt rgb "#00007F7F" pt 6 ps 0.5 lw 2 + + +# i 6 using ($5/SCALE) :2:(BAR_RELWD*($5/SCALE)) notitle with boxes lt rgb "grey" , \ +# '' i 8 using ($5/SCALE) :2:(BAR_RELWD*($5/SCALE)) notitle with boxes lt rgb "grey" , \ +# '' i 6 using ($5/SCALE) :($2-40):(BAR_RELWD*($5/SCALE)) notitle with boxes lt rgb "blue", \ +# '' i 8 using ($5/SCALE) :($2-40):(BAR_RELWD*($5/SCALE)) notitle with boxes lt rgb "blue", \ Index: doc/theses/mike_brooks_MMath/plots/string-allocn.py =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-allocn.py (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-allocn.py (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -31,5 +31,5 @@ 100:[-1.0, 0.1, 0.2, 0.5, 0.9], 200:[-1.0, 0.1, 0.2, 0.5, 0.9], - 500:[-1.0, 0.4, 0.9, 0.98]} + 500:[-1.0, 0.2, 0.4, 0.9, 0.98]} defaultExpansions = [-1, 0.2] @@ -56,5 +56,15 @@ ) -combined['is-default'] = np.isin(combined['expansion'], defaultExpansions).astype(int) +combined = combined.pivot_table( values=['op-duration-ns','hw_cur_req_mem(B)'], index=['corpus-meanlen-tgt', 'sut-platform', 'expansion'], aggfunc=['mean', 'min', 'max'] ) +combined = combined.reset_index() +combined.columns = combined.columns.to_flat_index() + +# text = combined.to_csv(header=True, index=True, sep='\t') +# print(text) + + +combined['is-default'] = np.isin(combined[('expansion','')], defaultExpansions).astype(int) + + # print ('!!') @@ -66,20 +76,27 @@ # First, for the CFA curves sut = "cfa" -sutGroup = combined.groupby('sut-platform').get_group(sut) +sutGroup = combined.groupby(('sut-platform','')).get_group(sut) -groupedSize = sutGroup.groupby('corpus-meanlen') +groupedSize = sutGroup.groupby(('corpus-meanlen-tgt','')) for sz, szgroup in groupedSize: if sz in favSizes.keys(): - szgroup_sorted = szgroup.sort_values(by='expansion') + szgroup_sorted = szgroup.sort_values(by=('expansion','')) print('"{sut}, len={len}"'.format(sut=sut, len=sz)) # print(szgroup_sorted) ## # print(szgroup_sorted['expansion'], 'isin', favSizes[sz]) ## - favoured = szgroup_sorted.loc[szgroup_sorted['expansion'].isin(favSizes[sz])] + favoured = szgroup_sorted.loc[szgroup_sorted[('expansion','')].isin(favSizes[sz])] # print('!') ## # print(favoured) ## - text = favoured[['expansion', 'op-duration-ns', 'hw_cur_req_mem(B)', 'is-default']].to_csv(header=False, index=False, sep='\t') + text = favoured[[('expansion',''), + ('mean','op-duration-ns'), + ('min','op-duration-ns'), + ('max','op-duration-ns'), + ('mean', 'hw_cur_req_mem(B)'), + ('min', 'hw_cur_req_mem(B)'), + ('max', 'hw_cur_req_mem(B)'), + 'is-default']].to_csv(header=False, index=False, sep='\t') print(text) print() @@ -89,9 +106,16 @@ atDefaults = combined.groupby('is-default').get_group(1) -for sz, szgroup in atDefaults.groupby('corpus-meanlen'): +for sz, szgroup in atDefaults.groupby(('corpus-meanlen-tgt','')): if sz in favSizes.keys(): print(sz) - text = szgroup[['expansion', 'op-duration-ns', 'hw_cur_req_mem(B)', 'sut-platform']].to_csv(header=False, index=False, sep='\t') + text = szgroup[[('expansion',''), + ('mean','op-duration-ns'), + ('min','op-duration-ns'), + ('max','op-duration-ns'), + ('mean', 'hw_cur_req_mem(B)'), + ('min', 'hw_cur_req_mem(B)'), + ('max', 'hw_cur_req_mem(B)'), + ('sut-platform','')]].to_csv(header=False, index=False, sep='\t') print(text) print() Index: doc/theses/mike_brooks_MMath/plots/string-pbv-fixcorp.py =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-pbv-fixcorp.py (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-pbv-fixcorp.py (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -26,9 +26,9 @@ -# Filter operation=pbv, corpus=1-*-1 +# Filter operation=pbv, corpus=100-*-*+*+l1 timings = timings.groupby('operation').get_group('pbv') -timings = timings.groupby('corpus-nstrs').get_group(1) -timings = timings.groupby('corpus-runid').get_group(1) +timings = timings.groupby('corpus-nstrs-tgt').get_group(100) +timings = timings.groupby('corpus-offset-instr').get_group('l1') @@ -41,8 +41,8 @@ if sut in prettyFieldNames: - sgroup_sorted = sgroup.sort_values(by='corpus-meanlen') + sgroup_sorted = sgroup.sort_values(by='corpusMeanLenCharsAct') print('"{header}"'.format(header=prettyFieldNames[sut])) - text = sgroup_sorted[['corpus-meanlen', 'op-duration-ns']].to_csv(header=False, index=False, sep='\t') + text = sgroup_sorted[['corpusMeanLenCharsAct', 'op-duration-ns']].to_csv(header=False, index=False, sep='\t') print(text) print() Index: doc/theses/mike_brooks_MMath/plots/string-pbv-varcorp.py =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-pbv-varcorp.py (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-pbv-varcorp.py (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -26,9 +26,9 @@ -# Filter operation=pbv, corpus=100-*-1 +# Filter operation=pbv, corpus=100-*-*+*+t0 timings = timings.groupby('operation').get_group('pbv') -timings = timings.groupby('corpus-nstrs').get_group(100) -timings = timings.groupby('corpus-runid').get_group(1) +timings = timings.groupby('corpus-nstrs-tgt').get_group(100) +timings = timings.groupby('corpus-offset-instr').get_group('t0') @@ -41,8 +41,8 @@ if sut in prettyFieldNames: - sgroup_sorted = sgroup.sort_values(by='corpus-meanlen') + sgroup_sorted = sgroup.sort_values(by='corpusMeanLenCharsAct') print('"{header}"'.format(header=prettyFieldNames[sut])) - text = sgroup_sorted[['corpus-meanlen', 'op-duration-ns']].to_csv(header=False, index=False, sep='\t') + text = sgroup_sorted[['corpusMeanLenCharsAct', 'op-duration-ns']].to_csv(header=False, index=False, sep='\t') print(text) print() Index: doc/theses/mike_brooks_MMath/plots/string-pbv.gp =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-pbv.gp (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-pbv.gp (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -25,12 +25,12 @@ set linetype 4 dashtype 2 plot INDIR."/plot-string-pbv-varcorp.dat" \ - i 0 using 1:2 title columnheader(1) with linespoints lt rgb "red" pt 3 ps 1 lw 1, \ - '' i 1 using 1:2 title columnheader(1) with linespoints lt rgb "blue" pt 6 ps 1 lw 1 + i 0 using 1:2 title columnheader(1) with points lt rgb "red" pt 3 ps 1, \ + '' i 1 using 1:2 title columnheader(1) with points lt rgb "blue" pt 6 ps 1 set xlabel "String length passed, fixed" set ylabel plot INDIR."/plot-string-pbv-fixcorp.dat" \ - i 0 using 1:2 title columnheader(1) with linespoints lt rgb "red" pt 3 ps 1 lw 1, \ - '' i 1 using 1:2 title columnheader(1) with linespoints lt rgb "blue" pt 6 ps 1 lw 1 + i 0 using 1:2 title columnheader(1) with points lt rgb "red" pt 3 ps 1, \ + '' i 1 using 1:2 title columnheader(1) with points lt rgb "blue" pt 6 ps 1 unset multiplot Index: doc/theses/mike_brooks_MMath/plots/string-peq-cppemu.gp =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-peq-cppemu.gp (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-peq-cppemu.gp (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -20,6 +20,6 @@ set linetype 4 dashtype 2 plot INDIR."/plot-string-peq-cppemu.dat" \ - i 0 using 1:2 title columnheader(1) with linespoints lt rgb "red" pt 2 ps 1 lw 1, \ - '' i 1 using 1:2 title columnheader(1) with linespoints lt rgb "red" pt 3 ps 1 lw 1, \ - '' i 2 using 1:2 title columnheader(1) with linespoints lt rgb "blue" pt 6 ps 1 lw 1, \ - '' i 3 using 1:2 title columnheader(1) with linespoints lt rgb "blue" pt 8 ps 1 lw 1 + i 0 using 1:2 title columnheader(1) with points lt rgb "red" pt 2 ps 1, \ + '' i 1 using 1:2 title columnheader(1) with points lt rgb "red" pt 1 ps 1, \ + '' i 2 using 1:2 title columnheader(1) with points lt rgb "blue" pt 6 ps 1, \ + '' i 3 using 1:2 title columnheader(1) with points lt rgb "blue" pt 8 ps 1 Index: doc/theses/mike_brooks_MMath/plots/string-peq-cppemu.py =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-peq-cppemu.py (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-peq-cppemu.py (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -27,9 +27,11 @@ timings = loadParseTimingData('result-append-pbv.csv') -# Filter operation=peq, corpus=100-*-1 +# Filter operation=peq, corpus=100-*-*+*+t0 timings = timings.groupby('operation').get_group('peq') -timings = timings.groupby('corpus-nstrs').get_group(100) -timings = timings.groupby('corpus-runid').get_group(1) +timings = timings.groupby('corpus-nstrs-tgt').get_group(100) +timings = timings.groupby('corpus-offset-instr').get_group('t0') + + # Emit in groups @@ -41,8 +43,8 @@ if sut in prettyFieldNames: - sgroup_sorted = sgroup.sort_values(by='corpus-meanlen') + sgroup_sorted = sgroup.sort_values(by='corpusMeanLenCharsAct') print('"{header}"'.format(header=prettyFieldNames[sut])) - text = sgroup_sorted[['corpus-meanlen', 'op-duration-ns']].to_csv(header=False, index=False, sep='\t') + text = sgroup_sorted[['corpusMeanLenCharsAct', 'op-duration-ns']].to_csv(header=False, index=False, sep='\t') print(text) print() Index: doc/theses/mike_brooks_MMath/plots/string-peq-sharing.gp =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-peq-sharing.gp (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-peq-sharing.gp (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -20,6 +20,6 @@ set linetype 4 dashtype 2 plot INDIR."/plot-string-peq-sharing.dat" \ - i 0 using 1:2 title columnheader(1) with linespoints lt rgb "red" pt 2 ps 1 lw 1, \ - '' i 1 using 1:2 title columnheader(1) with linespoints lt rgb "red" pt 3 ps 1 lw 1, \ - '' i 2 using 1:2 title columnheader(1) with linespoints lt rgb "blue" pt 6 ps 1 lw 1, \ - '' i 3 using 1:2 title columnheader(1) with linespoints lt rgb "blue" pt 8 ps 1 lw 1 + i 0 using 1:2 title columnheader(1) with points lt rgb "red" pt 2 ps 1, \ + '' i 1 using 1:2 title columnheader(1) with points lt rgb "red" pt 1 ps 1, \ + '' i 2 using 1:2 title columnheader(1) with points lt rgb "blue" pt 6 ps 1, \ + '' i 3 using 1:2 title columnheader(1) with points lt rgb "blue" pt 8 ps 1 Index: doc/theses/mike_brooks_MMath/plots/string-peq-sharing.py =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-peq-sharing.py (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-peq-sharing.py (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -28,9 +28,9 @@ -# Filter operation=peq, corpus=100-*-1 +# Filter operation=peq, corpus=100-*-*+*+t0 timings = timings.groupby('operation').get_group('peq') -timings = timings.groupby('corpus-nstrs').get_group(100) -timings = timings.groupby('corpus-runid').get_group(1) +timings = timings.groupby('corpus-nstrs-tgt').get_group(100) +timings = timings.groupby('corpus-offset-instr').get_group('t0') @@ -43,8 +43,8 @@ if sut in prettyFieldNames: - sgroup_sorted = sgroup.sort_values(by='corpus-meanlen') + sgroup_sorted = sgroup.sort_values(by='corpusMeanLenCharsAct') print('"{header}"'.format(header=prettyFieldNames[sut])) - text = sgroup_sorted[['corpus-meanlen', 'op-duration-ns']].to_csv(header=False, index=False, sep='\t') + text = sgroup_sorted[['corpusMeanLenCharsAct', 'op-duration-ns']].to_csv(header=False, index=False, sep='\t') print(text) print() Index: doc/theses/mike_brooks_MMath/plots/string-pta-sharing.gp =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-pta-sharing.gp (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-pta-sharing.gp (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -19,6 +19,6 @@ #show colornames plot INDIR."/plot-string-pta-sharing.dat" \ - i 0 using 1:2 title columnheader(1) with linespoints lt rgb "red" pt 2 ps 1 lw 1, \ - '' i 1 using 1:2 title columnheader(1) with linespoints lt rgb "dark-green" pt 4 ps 1 lw 1, \ - '' i 2 using 1:2 title columnheader(1) with linespoints lt rgb "blue" pt 6 ps 1 lw 1, \ - '' i 3 using 1:2 title columnheader(1) with linespoints lt rgb "dark-green" pt 12 ps 1 lw 1 + i 0 using 1:2 title columnheader(1) with points lt rgb "red" pt 2 ps 1, \ + '' i 1 using 1:2 title columnheader(1) with points lt rgb "dark-green" pt 4 ps 1, \ + '' i 2 using 1:2 title columnheader(1) with points lt rgb "blue" pt 6 ps 1, \ + '' i 3 using 1:2 title columnheader(1) with points lt rgb "dark-green" pt 12 ps 1 Index: doc/theses/mike_brooks_MMath/plots/string-pta-sharing.py =================================================================== --- doc/theses/mike_brooks_MMath/plots/string-pta-sharing.py (revision 241042418538c37dbb8067200a54e09c370946d2) +++ doc/theses/mike_brooks_MMath/plots/string-pta-sharing.py (revision f858ca51b2d9f59b5d8a2ff62b2a2c3403410a4e) @@ -32,8 +32,8 @@ -# Filter corpus=100-*-1 +# Filter corpus=100-*-1, corpus=100-*-*+*+t0 -timings = timings.groupby('corpus-nstrs').get_group(100) -timings = timings.groupby('corpus-runid').get_group(1) +timings = timings.groupby('corpus-nstrs-tgt').get_group(100) +timings = timings.groupby('corpus-offset-instr').get_group('t0') # Emit in groups @@ -50,8 +50,8 @@ if sut in opPretty: - sgroup_sorted = tgtOpTimings.sort_values(by='corpus-meanlen') + sgroup_sorted = tgtOpTimings.sort_values(by='corpusMeanLenCharsAct') print('"{header}"'.format(header=opPretty[sut])) - text = sgroup_sorted[['corpus-meanlen', 'op-duration-ns']].to_csv(header=False, index=False, sep='\t') + text = sgroup_sorted[['corpusMeanLenCharsAct', 'op-duration-ns']].to_csv(header=False, index=False, sep='\t') print(text) print()