ddidderr/zstd-rs
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge pull request #1225 from jennifermliu/dev

Browse Source 
Split samples when building dictionary for COVER
This commit is contained in:
Nick Terrell
2018-07-13 13:26:15 -07:00
committed by GitHub
parent aa2415402d 612b346ed5
commit 45821fac0c
8 changed files with 76 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files
programs/README.md
+1 -1
View File
@@ -150,7 +150,7 @@ Advanced arguments :
Dictionary builder :
--train ## : create a dictionary from a training set of files
--train-cover[=k=#,d=#,steps=#] : use the cover algorithm with optional args
--train-cover[=k=#,d=#,steps=#,split=#] : use the cover algorithm with optional args
--train-legacy[=s=#] : use the legacy algorithm with selectivity (default: 9)
-o file : `file` is dictionary name (default: dictionary)
--maxdict=# : limit dictionary to specified size (default: 112640)
programs/dibio.c
+2 -1
View File
@@ -323,7 +323,8 @@ int DiB_trainFromFiles(const char* dictFileName, unsigned maxDictSize,
srcBuffer, sampleSizes, fs.nbSamples,
coverParams);
if (!ZDICT_isError(dictSize)) {
DISPLAYLEVEL(2, "k=%u\nd=%u\nsteps=%u\n", coverParams->k, coverParams->d, coverParams->steps);
unsigned splitPercentage = (unsigned)(coverParams->splitPoint * 100);
DISPLAYLEVEL(2, "k=%u\nd=%u\nsteps=%u\nsplit=%u\n", coverParams->k, coverParams->d, coverParams->steps, splitPercentage);
}
} else {
dictSize = ZDICT_trainFromBuffer_cover(dictBuffer, maxDictSize, srcBuffer,
programs/zstd.1
+3 -2
View File
@@ -217,8 +217,9 @@ Split input files in blocks of size # (default: no split)
A dictionary ID is a locally unique ID that a decoder can use to verify it is using the right dictionary\. By default, zstd will create a 4\-bytes random number ID\. It\'s possible to give a precise number instead\. Short numbers have an advantage : an ID < 256 will only need 1 byte in the compressed frame header, and an ID < 65536 will only need 2 bytes\. This compares favorably to 4 bytes default\. However, it\'s up to the dictionary manager to not assign twice the same ID to 2 different dictionaries\.
.
.TP
\fB\-\-train\-cover[=k#,d=#,steps=#]\fR
Select parameters for the default dictionary builder algorithm named cover\. If \fId\fR is not specified, then it tries \fId\fR = 6 and \fId\fR = 8\. If \fIk\fR is not specified, then it tries \fIsteps\fR values in the range [50, 2000]\. If \fIsteps\fR is not specified, then the default value of 40 is used\. Requires that \fId\fR <= \fIk\fR\.
\fB\-\-train\-cover[=k#,d=#,steps=#,split=#]\fR
Select parameters for the default dictionary builder algorithm named cover\. If \fId\fR is not specified, then it tries \fId\fR = 6 and \fId\fR = 8\. If \fIk\fR is not specified, then it tries \fIsteps\fR values in the range [50, 2000]\. If \fIsteps\fR is not specified, then the default value of 40 is used\. If \fIsplit\fR is not specified or \fIsplit\fR <= 0, then the default value of 100 is used\. If \fIsplit\fR is 100, all input samples are used for both training and testing
to find optimal _d_ and _k_ to build dictionary.Requires that \fId\fR <= \fIk\fR\.
.
.IP
Selects segments of size \fIk\fR with highest score to put in the dictionary\. The score of a segment is computed by the sum of the frequencies of all the subsegments of size \fId\fR\. Generally \fId\fR should be in the range [6, 8], occasionally up to 16, but the algorithm will run faster with d <= \fI8\fR\. Good values for \fIk\fR vary widely based on the input data, but a safe range is [2 * \fId\fR, 2000]\. Supports multithreading if \fBzstd\fR is compiled with threading support\.
programs/zstd.1.md
+6 -1
View File
@@ -223,11 +223,12 @@ Compression of small files similar to the sample set will be greatly improved.
This compares favorably to 4 bytes default.
However, it's up to the dictionary manager to not assign twice the same ID to
2 different dictionaries.
* `--train-cover[=k#,d=#,steps=#]`:
* `--train-cover[=k#,d=#,steps=#,split=#]`:
Select parameters for the default dictionary builder algorithm named cover.
If _d_ is not specified, then it tries _d_ = 6 and _d_ = 8.
If _k_ is not specified, then it tries _steps_ values in the range [50, 2000].
If _steps_ is not specified, then the default value of 40 is used.
If _split_ is not specified or split <= 0, then the default value of 100 is used.
Requires that _d_ <= _k_.
Selects segments of size _k_ with highest score to put in the dictionary.
@@ -237,6 +238,8 @@ Compression of small files similar to the sample set will be greatly improved.
algorithm will run faster with d <= _8_.
Good values for _k_ vary widely based on the input data, but a safe range is
[2 * _d_, 2000].
If _split_ is 100, all input samples are used for both training and testing
to find optimal _d_ and _k_ to build dictionary.
Supports multithreading if `zstd` is compiled with threading support.
Examples:
@@ -249,6 +252,8 @@ Compression of small files similar to the sample set will be greatly improved.
`zstd --train-cover=k=50 FILEs`
`zstd --train-cover=k=50,split=60 FILEs`
* `--train-legacy[=selectivity=#]`:
Use legacy dictionary builder algorithm with the given dictionary
_selectivity_ (default: 9).
programs/zstdcli.c
+9 -2
View File
@@ -84,6 +84,7 @@ static U32 g_ldmMinMatch = 0;
static U32 g_ldmHashEveryLog = LDM_PARAM_DEFAULT;
static U32 g_ldmBucketSizeLog = LDM_PARAM_DEFAULT;
#define DEFAULT_SPLITPOINT 1.0
/*-************************************
* Display Macros
@@ -170,7 +171,7 @@ static int usage_advanced(const char* programName)
DISPLAY( "\n");
DISPLAY( "Dictionary builder : \n");
DISPLAY( "--train ## : create a dictionary from a training set of files \n");
DISPLAY( "--train-cover[=k=#,d=#,steps=#] : use the cover algorithm with optional args\n");
DISPLAY( "--train-cover[=k=#,d=#,steps=#,split=#] : use the cover algorithm with optional args\n");
DISPLAY( "--train-legacy[=s=#] : use the legacy algorithm with selectivity (default: %u)\n", g_defaultSelectivityLevel);
DISPLAY( " -o file : `file` is dictionary name (default: %s) \n", g_defaultDictName);
DISPLAY( "--maxdict=# : limit dictionary to specified size (default: %u) \n", g_defaultMaxDictSize);
@@ -282,10 +283,15 @@ static unsigned parseCoverParameters(const char* stringPtr, ZDICT_cover_params_t
if (longCommandWArg(&stringPtr, "k=")) { params->k = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; }
if (longCommandWArg(&stringPtr, "d=")) { params->d = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; }
if (longCommandWArg(&stringPtr, "steps=")) { params->steps = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; }
if (longCommandWArg(&stringPtr, "split=")) {
unsigned splitPercentage = readU32FromChar(&stringPtr);
params->splitPoint = (double)splitPercentage / 100.0;
if (stringPtr[0]==',') { stringPtr++; continue; } else break;
}
return 0;
}
if (stringPtr[0] != 0) return 0;
DISPLAYLEVEL(4, "cover: k=%u\nd=%u\nsteps=%u\n", params->k, params->d, params->steps);
DISPLAYLEVEL(4, "cover: k=%u\nd=%u\nsteps=%u\nsplit=%u\n", params->k, params->d, params->steps, (unsigned)(params->splitPoint * 100));
return 1;
}
@@ -310,6 +316,7 @@ static ZDICT_cover_params_t defaultCoverParams(void)
memset(&params, 0, sizeof(params));
params.d = 8;
params.steps = 4;
params.splitPoint = DEFAULT_SPLITPOINT;
return params;
}
#endif