Group CLI demo into subdirectory. (#6258)

CLI is not most developed interface. Putting them into correct directory can help new users to avoid it as most of the use cases are from a language binding.

demo/CLI/yearpredMSD/README.md

@@ -0,0 +1,9 @@
+Demonstrating how to use XGBoost on [Year Prediction task of Million Song Dataset](https://archive.ics.uci.edu/ml/datasets/YearPredictionMSD)
+
+1. Run runexp.sh
+```bash
+./runexp.sh
+```
+
+You can also use the script to prepare LIBSVM format, and run the [Distributed Version](../../multi-node).
+Note that though that normally you only need to use single machine for dataset at this scale, and use distributed version for larger scale dataset.

demo/CLI/yearpredMSD/csv2libsvm.py

@@ -0,0 +1,9 @@
+import sys
+fo = open(sys.argv[2], 'w')
+
+for l in open(sys.argv[1]):
+    arr = l.split(',')
+    fo.write('%s' % arr[0])
+    for i in range(len(arr) - 1):
+        fo.write(' %d:%s' % (i, arr[i+1]))
+fo.close()

demo/CLI/yearpredMSD/runexp.sh

@@ -0,0 +1,17 @@
+#!/bin/bash
+
+if [ -f YearPredictionMSD.txt ]
+then
+    echo "use existing data to run experiment"
+else
+    echo "getting data from uci, make sure you are connected to internet"
+    wget https://archive.ics.uci.edu/ml/machine-learning-databases/00203/YearPredictionMSD.txt.zip
+    unzip YearPredictionMSD.txt.zip
+fi
+echo "start making data.."
+# map feature using indicator encoding, also produce featmap.txt
+python csv2libsvm.py YearPredictionMSD.txt yearpredMSD.libsvm
+head -n 463715 yearpredMSD.libsvm > yearpredMSD.libsvm.train
+tail -n 51630 yearpredMSD.libsvm > yearpredMSD.libsvm.test
+echo "finish making the data"
+../../../xgboost yearpredMSD.conf

demo/CLI/yearpredMSD/yearpredMSD.conf

@@ -0,0 +1,29 @@
+# General Parameters, see comment for each definition
+# choose the tree booster, can also change to gblinear
+booster = gbtree
+# this is the only difference with classification, use reg:squarederror to do linear classification
+# when labels are in [0,1] we can also use reg:logistic
+objective = reg:squarederror
+
+# Tree Booster Parameters
+# step size shrinkage
+eta = 1.0
+# minimum loss reduction required to make a further partition
+gamma = 1.0
+# minimum sum of instance weight(hessian) needed in a child
+min_child_weight = 1
+# maximum depth of a tree
+max_depth = 5
+
+base_score = 2001
+# Task parameters
+# the number of round to do boosting
+num_round = 100
+# 0 means do not save any model except the final round model
+save_period = 0
+# The path of training data
+data = "yearpredMSD.libsvm.train"
+# The path of validation data, used to monitor training process, here [test] sets name of the validation set
+eval[test] = "yearpredMSD.libsvm.test"
+# The path of test data
+#test:data = "yearpredMSD.libsvm.test"