Merge branch 'master' of github.com:0xdata/h2o

py/h2o.py

@@ -649,7 +649,7 @@ def __url(self, loc, port=None):
         u = 'http://%s:%d/%s' % (self.http_addr, port, loc)
         return u 
 
-    def __do_json_request(self, jsonRequest=None, fullUrl=None, timeout=10, params={}, 
+    def __do_json_request(self, jsonRequest=None, fullUrl=None, timeout=10, params=None, 
         cmd='get', extraComment=None, ignoreH2oError=False, **kwargs):
         # if url param is used, use it as full url. otherwise crate from the jsonRequest
         if fullUrl:
@@ -659,12 +659,11 @@ def __do_json_request(self, jsonRequest=None, fullUrl=None, timeout=10, params={
 
         # remove any params that are 'None'
         # need to copy dictionary, since can't delete while iterating
-        params2 = params.copy()
-        for k in params2:
-            if params2[k] is None:
-                del params[k]
-
         if params is not None:
+            params2 = params.copy()
+            for k in params2:
+                if params2[k] is None:
+                    del params[k]
             paramsStr =  '?' + '&'.join(['%s=%s' % (k,v) for (k,v) in params.items()])
         else:
             paramsStr = ''

src/main/java/hex/LR2.java

@@ -52,8 +52,8 @@ public static class CalcSumsTask extends MRTask2<CalcSumsTask> {
     double _sumX,_sumY,_sumX2; // Sum of X's, Y's, X^2's
     @Override public void map( Chunk xs, Chunk ys ) {
       for( int i=0; i<xs._len; i++ ) {
-        double X = xs.getd(i);
-        double Y = ys.getd(i);
+        double X = xs.at0(i);
+        double Y = ys.at0(i);
         if( !Double.isNaN(X) && !Double.isNaN(Y)) {
           _sumX += X;
           _sumY += Y;
@@ -77,8 +77,8 @@ public static class CalcSquareErrorsTasks extends MRTask2<CalcSquareErrorsTasks>
     CalcSquareErrorsTasks( double meanX, double meanY ) { _meanX = meanX; _meanY = meanY; }
     @Override public void map( Chunk xs, Chunk ys ) {
       for( int i=0; i<xs._len; i++ ) {
-        double Xa = xs.getd(i);
-        double Ya = ys.getd(i);
+        double Xa = xs.at0(i);
+        double Ya = ys.at0(i);
         if(!Double.isNaN(Xa) && !Double.isNaN(Ya)) {
           Xa -= _meanX;
           Ya -= _meanY;
@@ -103,7 +103,7 @@ public static class CalcRegressionTask extends MRTask2<CalcRegressionTask> {
     CalcRegressionTask(double beta0, double beta1, double meanY) {_beta0=beta0; _beta1=beta1; _meanY=meanY;}
     @Override public void map( Chunk xs, Chunk ys ) {
       for( int i=0; i<xs._len; i++ ) {
-        double X = xs.getd(i); double Y = ys.getd(i);
+        double X = xs.at0(i); double Y = ys.at0(i);
         if( !Double.isNaN(X) && !Double.isNaN(Y) ) {
           double fit = _beta1*X + _beta0;
           double rs = fit-Y;

src/main/java/water/H2O.java

@@ -522,7 +522,64 @@ public static class OptArgs extends Arguments.Opt {
     public String random_udp_drop = null; // test only, randomly drop udp incoming
     public int pparse_limit = Integer.MAX_VALUE;
     public String no_requests_log = null; // disable logging of Web requests
+    public String h = null;
+    public String help = null;
   }
+
+  public static void printHelp() {
+    String s =
+    "Start an H2O node.\n" +
+    "\n" +
+    "Usage:  java [-Xmx<size>] -jar h2o.jar [options]\n" +
+    "        (Note that every option has a default and is optional.)\n" +
+    "\n" +
+    "    -name <h2oCloudName>\n" +
+    "          Cloud name used for Multicast discovery.\n" +
+    "          Nodes with the same cloud name will form an H2O cloud\n" +
+    "          (also known as an H2O cluster).\n" +
+    "          (Not to be used with -flatfile.)\n" +
+    "\n" +
+    "    -flatfile <flatFileName>\n" +
+    "          Configuration file explicitly listing H2O cloud node members.\n" +
+    "          (Not to be used with -name.)\n" +
+    "\n" +
+    "    -ip <ipAddressOfNode>\n" +
+    "          IP address of this node.\n" +
+    "\n" +
+    "    -port <port>\n" +
+    "          Port number for this node (note: port+1 is also used).\n" +
+    "          (The default port is " + DEFAULT_PORT + ".)\n" +
+    "\n" +
+    "    -ice_root <fileSystemPath>" +
+    "          The directory where H2O spills temporary data to disk." +
+    "          (The default is '" + DEFAULT_ICE_ROOT + "'.)\n" +
+    "\n" +
+    "    -h | -help\n" +
+    "          Print this help.\n" +
+    "\n" +
+    "Cloud formation behavior:\n" +
+    "\n" +
+    "    New H2O nodes join together to form a cloud at startup time.\n" +
+    "    Once a cloud is given work to perform, it locks out new members.\n" +
+    "    from joining.\n" +
+    "\n" +
+    "Examples:\n" +
+    "\n" +
+    "    Start an H2O node with 4GB of memory and a default cloud name:\n" +
+    "        java -Xmx4g -jar h2o.jar\n" +
+    "\n" +
+    "    Start an H2O node with 6GB of memory and a specify the cloud name:\n" +
+    "        java -Xmx6g -jar h2o.jar -name MyCloud\n" +
+    "\n" +
+    "    Start an H2O cloud with three 2GB nodes and a default cloud name:\n" +
+    "        java -Xmx2g -jar h2o.jar\n" +
+    "        java -Xmx2g -jar h2o.jar\n" +
+    "        java -Xmx2g -jar h2o.jar\n" +
+    "\n";
+
+    System.out.print(s);
+  }
+
   public static boolean IS_SYSTEM_RUNNING = false;
 
   // Start up an H2O Node and join any local Cloud
@@ -536,6 +593,12 @@ public static void main( String[] args ) {
     Arguments arguments = new Arguments(args);
     arguments.extract(OPT_ARGS);
     ARGS = arguments.toStringArray();
+
+    if ((OPT_ARGS.h != null) || (OPT_ARGS.help != null)) {
+      printHelp();
+      System.exit (0);
+    }
+
     ParseDataset.PLIMIT = OPT_ARGS.pparse_limit;
 
     // Get ice path before loading Log or Persist class
@@ -546,6 +609,7 @@ public static void main( String[] args ) {
     } catch(URISyntaxException ex) {
       throw new RuntimeException("Invalid ice_root: " + ice + ", " + ex.getMessage());
     }
+    Log.info ("ICE root: '" + ICE_ROOT + "'");
 
     SELF_ADDRESS = findInetAddressForSelf();
 
@@ -592,6 +656,8 @@ private static void startLocalNode() {
       STATIC_H2OS.add(SELF);
     }
 
+    Log.info ("H2O cloud name: '" + NAME + "'");
+
     Log.info("(v"+VERSION+") '"+NAME+"' on " + SELF+(OPT_ARGS.flatfile==null
         ? (", discovery address "+CLOUD_MULTICAST_GROUP+":"+CLOUD_MULTICAST_PORT)
             : ", static configuration based on -flatfile "+OPT_ARGS.flatfile));

src/samples/water/Part05_KMeansNewAPI.java

@@ -35,8 +35,10 @@ public static void main(String[] args) throws Exception {
       Random rand = new Random();
       for( int cluster = 0; cluster < clusters.length; cluster++ ) {
         long row = Math.max(0, (long) (rand.nextDouble() * frame.vecs().length) - 1);
-        for( int i = 0; i < frame.vecs().length; i++ )
-          clusters[cluster][i] = frame.vecs()[i].getd(row);
+        for( int i = 0; i < frame.vecs().length; i++ ) {
+          Vec v = frame.vecs()[i];
+          clusters[cluster][i] = v.at(row);
+        }
       }
 
       // Iterate over the dataset and show error for each step
@@ -89,7 +91,7 @@ public static class KMeans extends MRTask2<KMeans> {
         for( int cluster = 0; cluster < _clusters.length; cluster++ ) {
           double sqr = 0;           // Sum of dimensional distances
           for( int column = 0; column < chunks.length; column++ ) {
-            double delta = chunks[column].atd(row) - _clusters[cluster][column];
+            double delta = chunks[column].at0(row) - _clusters[cluster][column];
             sqr += delta * delta;
           }
           if( sqr < minSqr ) {
@@ -101,7 +103,7 @@ public static class KMeans extends MRTask2<KMeans> {
 
         // Add values and increment counter for chosen cluster
         for( int column = 0; column < chunks.length; column++ )
-          _sums[nearest][column] += chunks[column].atd(row);
+          _sums[nearest][column] += chunks[column].at0(row);
         _counts[nearest]++;
       }
       _clusters = null;