Natools

Check-in [6877c806ab]
Login
☰Home Timeline Files Branches Tags Tickets Wiki Login
Overview
Comment:tools/smaz: add selection of the scoring method in optimization
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: 6877c806ab8df40620ed6f27a8fd33157359e5d8
User & Date: nat on 2016-10-31 22:53:58
Other Links: manifest | tags
Context
2016-11-01
21:32
tools/smaz: fix option arguments in help text check-in: 8a29fd3a72 user: nat tags: trunk
2016-10-31
22:53
tools/smaz: add selection of the scoring method in optimization check-in: 6877c806ab user: nat tags: trunk
2016-10-30
18:19
tools/sxcat: add a command-line option to output a list of input atoms check-in: ac88f5abfb user: nat tags: trunk
Changes
Hide Diffs Side-by-Side Diffs Ignore Whitespace Patch

Modified tools/smaz.adb from [fa5ab81891] to [cc397d7cf7].

︙
52
53
54
55
56
57
58




59
60
61
62
63
64
65

52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69








+
+
+
+









   package Dict_Sources is
      type Enum is
        (S_Expression,
         Text_List,
         Unoptimized_Text_List);
   end Dict_Sources;

   package Methods is
      type Enum is (Encoded, Frequency, Gain);
   end Methods;

   package Options is
      type Id is
        (Output_Ada_Dict,
         Dictionary_Input,
         Decode,
         Encode,
︙
73
74
75
76
77
78
79
80


81
82
83
84
85
86
87
88
89
90
91
92
93
94
95

96
97
98
99
100
101
102

77
78
79
80
81
82
83

84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108








-
+
+















+








         Max_Sub_Size,
         Stat_Output,
         No_Stat_Output,
         Text_List_Input,
         Fast_Text_Input,
         Max_Word_Size,
         Sx_Output,
         No_Sx_Output);
         No_Sx_Output,
         Score_Method);
   end Options;

   package Getopt is new Natools.Getopt_Long (Options.Id);

   type Callback is new Getopt.Handlers.Callback with record
      Display_Help : Boolean := False;
      Need_Dictionary : Boolean := False;
      Stat_Output : Boolean := False;
      Sx_Output : Boolean := False;
      Sx_Dict_Output : Boolean := False;
      Min_Sub_Size : Positive := 1;
      Max_Sub_Size : Positive := 3;
      Max_Word_Size : Positive := 10;
      Job_Count : Natural := 0;
      Filter_Threshold : Natools.Smaz.Tools.String_Count := 0;
      Score_Method : Methods.Enum := Methods.Encoded;
      Action : Actions.Enum := Actions.Nothing;
      Ada_Dictionary : Ada.Strings.Unbounded.Unbounded_String;
      Hash_Package : Ada.Strings.Unbounded.Unbounded_String;
      Dict_Source : Dict_Sources.Enum := Dict_Sources.S_Expression;
   end record;

   overriding procedure Option
︙
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142












143
144
145
146
147


148
149
150
151
152
153
154

124
125
126
127
128
129
130











131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153

154
155
156
157
158
159
160
161
162








-
-
-
-
-
-
-
-
-
-
-







+
+
+
+
+
+
+
+
+
+
+
+




-
+
+








      Counts : out Natools.Smaz.Tools.Dictionary_Counts);
      --  Dispatch to parallel or non-parallel version of Evaluate_Dictionary
      --  depending on Job_Count.

   function Getopt_Config return Getopt.Configuration;
      --  Build the configuration object

   procedure Optimization_Round
     (Dict : in out Holders.Holder;
      Score : in out Ada.Streams.Stream_Element_Count;
      Counts : in out Natools.Smaz.Tools.Dictionary_Counts;
      Pending_Words : in out Natools.Smaz.Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz.Tools.String_Lists.List;
      Job_Count : in Natural;
      Updated : out Boolean);
      --  Try to improve on Dict by replacing a single entry from it with
      --  one of the substring in Pending_Words.

   function Length
     (Dictionary : in Natools.Smaz.Dictionary;
      E : in Ada.Streams.Stream_Element)
     return Score_Value
     is (Natools.Smaz.Dict_Entry (Dictionary, E)'Length);
      --  Length of a dictionary entry

   procedure Optimization_Round
     (Dict : in out Holders.Holder;
      Score : in out Ada.Streams.Stream_Element_Count;
      Counts : in out Natools.Smaz.Tools.Dictionary_Counts;
      Pending_Words : in out Natools.Smaz.Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz.Tools.String_Lists.List;
      Job_Count : in Natural;
      Method : in Methods.Enum;
      Updated : out Boolean);
      --  Try to improve on Dict by replacing a single entry from it with
      --  one of the substring in Pending_Words.

   function Optimize_Dictionary
     (Base : in Natools.Smaz.Dictionary;
      Pending_Words : in Natools.Smaz.Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz.Tools.String_Lists.List;
      Job_Count : in Natural)
      Job_Count : in Natural;
      Method : in Methods.Enum)
     return Natools.Smaz.Dictionary;
      --  Optimize the dictionary on Input_Texts, starting with Base and
      --  adding substrings from Pending_Words.

   procedure Parallel_Evaluate_Dictionary
     (Job_Count : in Positive;
      Dict : in Natools.Smaz.Dictionary;
︙
193
194
195
196
197
198
199












200
201
202
203
204
205
206
207
208


209
210
211
212
213
214
215

201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227

228
229
230
231
232
233
234
235
236








+
+
+
+
+
+
+
+
+
+
+
+








-
+
+








     (Dictionary : in Natools.Smaz.Dictionary;
      Counts : in Natools.Smaz.Tools.Dictionary_Counts;
      E : Ada.Streams.Stream_Element)
     return Score_Value
     is (Score_Value (Counts (E)) * (Length (Dictionary, E) - 1));
      --  Score value using the number of bytes saved using E

   function Score
     (Dictionary : in Natools.Smaz.Dictionary;
      Counts : in Natools.Smaz.Tools.Dictionary_Counts;
      E : in Ada.Streams.Stream_Element;
      Method : in Methods.Enum)
     return Score_Value
     is (case Method is
         when Methods.Encoded => Score_Encoded (Dictionary, Counts, E),
         when Methods.Frequency => Score_Frequency (Dictionary, Counts, E),
         when Methods.Gain => Score_Gain (Dictionary, Counts, E));
      --  Scare value with dynamically chosen method

   function To_Dictionary
     (Handler : in Callback'Class;
      Input : in Natools.Smaz.Tools.String_Lists.List)
     return Natools.Smaz.Dictionary;
      --  Convert the input into a dictionary given the option in Handler

   function Worst_Index
     (Dict : in Natools.Smaz.Dictionary;
      Counts : in Natools.Smaz.Tools.Dictionary_Counts)
      Counts : in Natools.Smaz.Tools.Dictionary_Counts;
      Method : in Methods.Enum)
     return Ada.Streams.Stream_Element;
      --  Remove the worstly-scored item from Dict


   overriding procedure Option
     (Handler  : in out Callback;
      Id       : in Options.Id;
︙
283
284
285
286
287
288
289



290
291
292
293
294
295
296

304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320








+
+
+









         when Options.Job_Count =>
            Handler.Job_Count := Natural'Value (Argument);

         when Options.Filter_Threshold =>
            Handler.Filter_Threshold
              := Natools.Smaz.Tools.String_Count'Value (Argument);

         when Options.Score_Method =>
            Handler.Score_Method := Methods.Enum'Value (Argument);
      end case;
   end Option;


   procedure Evaluate_Dictionary
     (Job_Count : in Natural;
      Dict : in Natools.Smaz.Dictionary;
︙
347
348
349
350
351
352
353

354
355
356
357
358
359
360
361
362
363
364
365

366
367
368
369
370
371
372
373

374
375
376
377
378
379
380

371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398

399
400
401
402
403
404
405
406








+












+







-
+








      R.Add_Option ("stats",         's', No_Argument,       Stat_Output);
      R.Add_Option ("no-stats",      'S', No_Argument,       No_Stat_Output);
      R.Add_Option ("text-list",     't', No_Argument,       Text_List_Input);
      R.Add_Option ("fast-text-list", 'T', No_Argument,       Fast_Text_Input);
      R.Add_Option ("max-word-len",  'W', Required_Argument, Max_Word_Size);
      R.Add_Option ("s-expr",        'x', No_Argument,       Sx_Output);
      R.Add_Option ("no-s-expr",     'X', No_Argument,       No_Sx_Output);
      R.Add_Option ("score-method",       Required_Argument, Score_Method);

      return R;
   end Getopt_Config;


   procedure Optimization_Round
     (Dict : in out Holders.Holder;
      Score : in out Ada.Streams.Stream_Element_Count;
      Counts : in out Natools.Smaz.Tools.Dictionary_Counts;
      Pending_Words : in out Natools.Smaz.Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz.Tools.String_Lists.List;
      Job_Count : in Natural;
      Method : in Methods.Enum;
      Updated : out Boolean)
   is
      use type Ada.Streams.Stream_Element_Offset;

      New_Value : Ada.Strings.Unbounded.Unbounded_String;
      New_Position : Natools.Smaz.Tools.String_Lists.Cursor;
      Worst_Index : constant Ada.Streams.Stream_Element
        := Smaz.Worst_Index (Dict.Element, Counts);
        := Smaz.Worst_Index (Dict.Element, Counts, Method);
      Worst_Value : constant String
        := Natools.Smaz.Dict_Entry (Dict.Element, Worst_Index);
      Worst_Count : constant Natools.Smaz.Tools.String_Count
        := Counts (Worst_Index);
      Base : constant Natools.Smaz.Dictionary
        := Natools.Smaz.Tools.Remove_Element (Dict.Element, Worst_Index);
      Old_Score : constant Ada.Streams.Stream_Element_Count := Score;
︙
426
427
428
429
430
431
432
433


434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451

452
453
454
455
456
457
458

452
453
454
455
456
457
458

459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486








-
+
+


















+








   end Optimization_Round;


   function Optimize_Dictionary
     (Base : in Natools.Smaz.Dictionary;
      Pending_Words : in Natools.Smaz.Tools.String_Lists.List;
      Input_Texts : in Natools.Smaz.Tools.String_Lists.List;
      Job_Count : in Natural)
      Job_Count : in Natural;
      Method : in Methods.Enum)
     return Natools.Smaz.Dictionary
   is
      Holder : Holders.Holder := Holders.To_Holder (Base);
      Pending : Natools.Smaz.Tools.String_Lists.List := Pending_Words;
      Score : Ada.Streams.Stream_Element_Count;
      Counts : Natools.Smaz.Tools.Dictionary_Counts;
      Running : Boolean := True;
   begin
      Evaluate_Dictionary (Job_Count, Base, Input_Texts, Score, Counts);

      while Running loop
         Optimization_Round
           (Holder,
            Score,
            Counts,
            Pending,
            Input_Texts,
            Job_Count,
            Method,
            Running);
      end loop;

      return Holder.Element;
   end Optimize_Dictionary;
︙
702
703
704
705
706
707
708






709
710
711
712
713
714
715

730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749








+
+
+
+
+
+









            when Options.Filter_Threshold =>
               Put_Line (Output, " <threshold>");
               Put_Line (Output, Indent & Indent
                 & "Before building a dictionary from substrings, remove");
               Put_Line (Output, Indent & Indent
                 & "substrings whose count is below the threshold.");

            when Options.Score_Method =>
               Put_Line (Output, " <method>");
               Put_Line (Output, Indent & Indent
                 & "Select heuristic method to replace dictionary items"
                 & " during optimization");
         end case;
      end loop;
   end Print_Help;


   function To_Dictionary
     (Handler : in Callback'Class;
︙
750
751
752
753
754
755
756
757


758
759
760
761
762
763
764
765
766
767
768
769
770
771


772
773
774
775
776
777
778
779

780
781
782
783
784
785
786

784
785
786
787
788
789
790

791
792
793
794
795
796
797
798
799
800
801
802
803
804
805

806
807
808
809
810
811
812
813
814

815
816
817
818
819
820
821
822








-
+
+













-
+
+







-
+








                     Natools.Smaz.Tools.Simple_Dictionary_And_Pending
                       (Counter, 254, Selected, Pending);

                     return Optimize_Dictionary
                       (Natools.Smaz.Tools.To_Dictionary (Selected, True),
                        Pending,
                        Input,
                        Handler.Job_Count);
                        Handler.Job_Count,
                        Handler.Score_Method);
                  end;
               else
                  return Natools.Smaz.Tools.To_Dictionary
                    (Natools.Smaz.Tools.Simple_Dictionary (Counter, 254),
                     True);
               end if;
            end;
      end case;
   end To_Dictionary;


   function Worst_Index
     (Dict : in Natools.Smaz.Dictionary;
      Counts : in Natools.Smaz.Tools.Dictionary_Counts)
      Counts : in Natools.Smaz.Tools.Dictionary_Counts;
      Method : in Methods.Enum)
     return Ada.Streams.Stream_Element
   is
      Result : Ada.Streams.Stream_Element := 0;
      Worst_Score : Score_Value := Score_Encoded (Dict, Counts, 0);
      S : Score_Value;
   begin
      for I in 1 .. Dict.Dict_Last loop
         S := Score_Encoded (Dict, Counts, I);
         S := Score (Dict, Counts, I, Method);

         if S < Worst_Score then
            Result := I;
            Worst_Score := S;
         end if;
      end loop;
︙

This page was generated in about 0.006s by Fossil 2.23 [47362306a7] 2023-11-01 18:56:47