Changeset e0350e0 for doc/theses/mike_brooks_MMath/plots/common.py

Timestamp:

Jun 4, 2025, 1:43:34 PM (4 months ago)

Author:

Michael Brooks <mlbrooks@…>

Branches:

master

Children:

f858ca5

Parents:

c8bdbaf

Message:

Recent rework of string benchmarks

File:

• doc/theses/mike_brooks_MMath/plots/common.py (modified) (2 diffs)

doc/theses/mike_brooks_MMath/plots/common.py

@@ -17 +17 @@
 
     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 +35 @@
     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,