Changeset 19802 for trunk/FACT++/src/root2csv.cc

Timestamp:

10/27/19 11:11:01 (5 years ago)

Author:

tbretz

Message:

Moved the common Splitting algorithm to a common class 'Splitting' (added) and added the possibility to rename columns (csv2root)

File:

• trunk/FACT++/src/root2csv.cc (modified) (12 diffs)

trunk/FACT++/src/root2csv.cc

@@ -1 @@
-#include <random>
-
 #include <boost/regex.hpp>
 #include <boost/filesystem.hpp>
@@ -7 +5 @@
 #include "tools.h"
 #include "Time.h"
-#include "Configuration.h"
+#include "Splitting.h"
 
 #include <TROOT.h>
@@ -65 +63 @@
         ;
 
-    po::options_description split("Splitting options");
-    split.add_options()
-        ("split-sequence,S", vars<uint16_t>(),            "Split data sequentially into several trees/files (e.g. 1, 1, 2)")
-        ("split-quantile,Q", vars<double>(),              "Split data randomly into several trees/files (e.g. 0.5, 1)")
-        ("seed", var<uint64_t>(mt19937_64::default_seed), "Seed value in case of random split")
-        ;
-
     po::options_description debug("Debug options");
     debug.add_options()
@@ -84 +75 @@
 
     conf.AddOptions(control);
-    conf.AddOptions(split);
+    conf.AddOptions(Tools::Splitting::options());
     conf.AddOptions(debug);
     conf.SetArgumentPositions(p);
@@ -169 +160 @@
         "and all negative values are considered 'fales' (discard the entry).\n"
         "\n"
-        "For several purposes, it might be convenient to split the output to several "
-        "files. This can be achieved using the --split-sequence (-S) "
-        "and the --split-quantile (-Q) options. If a split sequence is defined as "
-        "-S 1 -S 2 -S 1 the events are split by 1:2:1 in this sequence order. If "
-        "quantiles are given as -Q 0.5 -Q 0.6, the first tree will contain 50% of "
-        "the second one 10% and the third one 40%. The corresponding seed value can "
-        "be set with --seed. Filenames are then created by adding an index after(!) "
-        "the extension, e.g. file.csv-0, file.csv-1, ...\n"
+        << Tools::Splitting::usage() <<
         "\n"
         "In case of success, 0 is returned, a value>0 otherwise.\n"
@@ -403 +387 @@
     const vector<Map> autoalias  = conf.Vec<Map>("auto-alias");
 
-    // ----------------------------- Setup splitting ---------------------------
-
-    vector<uint16_t> split_seq   = conf.Vec<uint16_t>("split-sequence");
-    vector<double>   split_quant = conf.Vec<double>("split-quantile");
-
-    if (!split_seq.empty() && !split_quant.empty())
-        throw runtime_error("Only splitting by --split-sequence or --split-quantile is allowed.");
-
-    const size_t num_split = split_seq.size()+split_quant.size()==0 ? 0 :
-        ::max(split_seq.size(), split_quant.size()+1);
-
-    map<size_t, size_t> split_lut;
-    for (size_t i=0; i<split_seq.size(); i++)
-    {
-        const size_t sz = split_lut.size();
-        for (size_t j=0; j<split_seq[i]; j++)
-            split_lut.emplace(j+sz, i);
-    }
-
-    for (size_t i=0; i<split_quant.size(); i++)
-        if (split_quant[i]<0 || split_quant[i]>=1)
-            throw runtime_error("Splitting quantiles must be in the range [0;1)");
-
-    for (size_t i=1; i<split_quant.size(); i++)
-    {
-        if (split_quant[i]<=split_quant[i-1])
-            throw runtime_error("Splitting quantiles must be in increasing order.");
-    }
-
     // -------------------------------------------------------------------------
 
-    const uniform_real_distribution<double> distribution(0,1);
-    mt19937_64 generator;
-    generator.seed(conf.Get<uint64_t>("seed"));
-    auto rndm = bind(distribution, generator);
-
-    // -------------------------------------------------------------------------
+    /*const*/ Tools::Splitting split(conf);
 
     if (verbose>0)
@@ -780 +730 @@
     // -------------------------------------------------------------------------
 
-    if (num_split)
-    {
-        cout << "\nSplitting configured " << (split_seq.empty()?"randomly":"in sequence") << " into " << num_split << " files." << endl;
-        if (!split_quant.empty())
-            cout << "Seed value configured as " << conf.Get<uint64_t>("seed") << "." << endl;
+    if (verbose>0)
+    {
+        cout << '\n';
+        split.print();
     }
 
@@ -798 +747 @@
     vector<ofstream> outfiles;
 
-    if (num_split==0)
+    if (split,empty())
     {
         TString path(out.c_str());
@@ -811 +760 @@
     else
     {
-        for (size_t i=0; i<num_split; i++)
+        for (size_t i=0; i<split.size(); i++)
         {
             TString path(out.c_str());
@@ -828 +777 @@
     // ---------------------------- Write Body --------------------------------
     size_t count = 0;
-    vector<size_t> ncount(num_split?num_split:1);
+    vector<size_t> ncount(split.empty()?1:split.size());
 
     auto itree = c.GetTreeNumber();
@@ -848 +797 @@
             continue;
 
-        size_t index = 0;
-        if (!split_lut.empty())
-            index = split_lut[count % split_lut.size()];
-        if (!split_quant.empty())
-        {
-            const float r = rndm();
-            for (; r>=split_quant[index]; index++)
-                if (index==split_quant.size())
-                    break;
-        }
+        const size_t index = split.index(count++);
+        ncount[index]++;
 
         vector<string> join;
@@ -875 +816 @@
 
         outfiles[index] << boost::join(join, " ") << "\n";
-
-        count ++;
-        ncount[index] ++;
     }