Java Frame.numCols方法代码示例

import water.fvec.Frame; //导入方法依赖的package包/类
/** Initialize the ModelBuilder, validating all arguments and preparing the
 *  training frame.  This call is expected to be overridden in the subclasses
 *  and each subclass will start with "super.init();".
 *
 *  Validate K, max_iterations and the number of rows. */
@Override public void init(boolean expensive) {
  super.init(expensive);
  if( _parms._max_iterations < 0 || _parms._max_iterations > 1e6) error("_max_iterations", " max_iterations must be between 0 and 1e6");
  if( _train == null ) return;
  if( _parms._init == Initialization.User && _parms._user_points == null )
    error("_user_y","Must specify initial cluster centers");
  if( null != _parms._user_points ){ // Check dimensions of user-specified centers
    Frame user_points = _parms._user_points.get();
    if( user_points.numCols() != _train.numCols() - numSpecialCols()) {
      error("_user_y","The user-specified points must have the same number of columns (" + (_train.numCols() - numSpecialCols()) + ") as the training observations");
    } else if( user_points.numRows() != _parms._k)
      error("_user_y","The number of rows in the user-specified points is not equal to k = " + _parms._k);
  }
  if (expensive && error_count() == 0) checkMemoryFootPrint();
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
@Override protected Frame transformImpl(Frame f) {
  ((ASTExec)_ast._asts[1])._asts[1] = AST.newASTFrame(f);
  Frame fr = Exec.execute(_ast).getFrame();
  _newCol = _newNames==null?new String[fr.numCols()]:_newNames;
  _newColTypes = toJavaPrimitive(fr.anyVec().get_type_str());
  if( (_multiColReturn=fr.numCols() > 1) ) {
    for(int i=0;i<_newCol.length;i++) {
      if(_newNames==null) _newCol[i] = f.uniquify(i > 0 ? _newCol[i - 1] : _oldCol);
      f.add(_newCol[i], fr.vec(i));
    }
    if( _inplace ) f.remove(f.find(_oldCol)).remove();
  } else {
    _newCol = new String[]{_inplace ? _oldCol : f.uniquify(_oldCol)};
    if( _inplace ) f.replace(f.find(_oldCol), fr.anyVec()).remove();
    else          f.add(_newNames==null?_newCol[0]:_newNames[0], fr.anyVec());
  }
  DKV.put(f);
  return f;
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
@Override Val apply( Env env, Env.StackHelp stk, AST asts[] ) {
  Frame frx = stk.track(asts[1].exec(env)).getFrame();
  Frame fry = stk.track(asts[2].exec(env)).getFrame();
  if( frx.numRows() != fry.numRows() )
    throw new IllegalArgumentException("Frames must have the same number of rows, found "+frx.numRows()+" and "+fry.numRows());
  if( frx.numCols() != fry.numCols() )
    throw new IllegalArgumentException("Frames must have the same number of columns, found "+frx.numCols()+" and "+fry.numCols());

  String use = stk.track(asts[3].exec(env)).getStr();
  Mode mode;
  switch( use ) {
  case "everything":            mode = Mode.Everything; break;
  case "all.obs":               mode = Mode.AllObs; break;
  case "complete.obs":          mode = Mode.CompleteObs; break;
  default: throw new IllegalArgumentException("unknown use mode, found: "+use);
  }

  return frx.numRows() == 1 ? scalar(frx,fry,mode) : array(frx,fry,mode);
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
/**
 * Helper to check parsed column names
 */
private void testColNames(String[] dataset, String[] exp, int len, String sep) {
  StringBuilder sb1 = new StringBuilder();
  for (String ds : dataset) sb1.append(ds).append(sep);
  Key k1 = makeByteVec(sb1.toString());
  Key r1 = Key.make("r1");
  ParseDataset.parse(r1, k1);
  Frame fr = DKV.get(r1).get();
  try {
    Assert.assertEquals(len, fr.numRows());
    Assert.assertEquals(exp.length, fr.numCols());
    for (int j = 0; j < fr.numCols(); ++j) {
      Assert.assertTrue(exp[j].equals(fr.names()[j]));
    }
  } finally {
    fr.delete();
  }
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
@Override Val apply(Env env, Env.StackHelp stk, AST asts[]) {
  Frame fr = stk.track(asts[1].exec(env)).getFrame();
  Vec v;
  if( fr.numCols()!=1 )
    throw new IllegalArgumentException("Unique applies to a single column only.");
  if( fr.anyVec().isCategorical() ) {
    v = Vec.makeSeq(0, (long)fr.anyVec().domain().length, true);
    v.setDomain(fr.anyVec().domain());
    DKV.put(v);
  } else {
    UniqTask t = new UniqTask().doAll(fr);
    int nUniq = t._uniq.size();
    final ASTGroup.G[] uniq = t._uniq.keySet().toArray(new ASTGroup.G[nUniq]);
    v = Vec.makeZero(nUniq);
    new MRTask() {
      @Override
      public void map(Chunk c) {
        int start = (int) c.start();
        for (int i = 0; i < c._len; ++i) c.set(i, uniq[i + start]._gs[0]);
      }
    }.doAll(v);
  }
  return new ValFrame(new Frame(v));
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
@Override Val apply( Env env, Env.StackHelp stk, AST asts[] ) {
  Val val = asts[1].exec(env);
  switch( val.type() ) {
  case Val.NUM: 
    double d = val.getNum();
    return new ValNum(Double.isNaN(d) ? d : op(new MutableDateTime(0),d));
  case Val.FRM: 
    Frame fr = stk.track(val).getFrame();
    if( fr.numCols() > 1 ) throw water.H2O.unimpl();
    return new ValFrame(new MRTask() {
        @Override public void map( Chunk chk, NewChunk cres ) {
          MutableDateTime mdt = new MutableDateTime(0,ParseTime.getTimezone());
          for( int i=0; i<chk._len; i++ )
            cres.addNum(chk.isNA(i) ? Double.NaN : op(mdt,chk.at8(i)));
        }
      }.doAll_numericResult(1,fr).outputFrame(fr._names, factors()));
  default: throw water.H2O.fail();
  }
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
public final boolean hasClosedForm() {
  long na_cnt = 0;
  Frame train = _train.get();
  for(int i = 0; i < train.numCols(); i++)
    na_cnt += train.vec(i).naCnt();
  return hasClosedForm(na_cnt);
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
FrameSynopsisV3(Frame fr) {
  Vec[] vecs = fr.vecs();

  frame_id = new FrameKeyV3(fr._key);
  _fr = fr;
  rows = fr.numRows();
  columns = vecs.length;
  byte_size = fr.byteSize();
  is_text = fr.numCols()==1 && vecs[0] instanceof ByteVec;
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
public static double sparseRatio(Frame fr) {
  double reg = 1.0/fr.numCols();
  double res = 0;
  for(Vec v:fr.vecs())
    res += v.sparseRatio();
  return res * reg;
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
public static void initCheck(Frame input, int width, int height, int depth) {
  ConcurrencyUtils.setNumberOfThreads(1);
  if (width < 1 || height < 1 || depth < 1)
    throw new H2OIllegalArgumentException("dimensions must be >= 1");
  if (width*height*depth != input.numCols())
    throw new H2OIllegalArgumentException("dimensions HxWxD must match the # columns of the frame");
  for (Vec v : input.vecs()) {
    if (v.naCnt() > 0)
      throw new H2OIllegalArgumentException("DCT can not be computed on rows with missing values");
    if (!v.isNumeric())
      throw new H2OIllegalArgumentException("DCT can only be computed on numeric columns");
  }
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
@Override public Transform<H2OScaler> fit(Frame f) {
  means = new double[f.numCols()];
  sdevs = new double[f.numCols()];
  for(int i=0;i<f.numCols();++i) {
    means[i] = f.vec(i).mean();
    sdevs[i] = f.vec(i).sigma();
  }
  return this;
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
@Override Val apply( Env env, Env.StackHelp stk, AST asts[] ) {
  Frame ary = stk.track(asts[1].exec(env)).getFrame();
  if( ary.numCols() != 1 ) throw new IllegalArgumentException("character requires a single column");
  Vec v0 = ary.anyVec();
  Vec v1 = v0.isString() ? null : v0.toStringVec(); // toCategoricalVec() creates a new vec --> must be cleaned up!
  Frame fr = new Frame(ary._names, new Vec[]{v1 == null ? v0.makeCopy(null) : v1});
  return new ValFrame(fr);
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
/**
 * Transpose the Frame as if it was a matrix (rows <-> columns).
 * Must be all numeric, will fail if there are too many rows ( >= ~.5M).
 *
 * Result is made to be compatible (i.e. the same vector group and chunking) with the target frame.
 *
 * @param src
 * @return
 */
public static Frame transpose(Frame src, Frame tgt){
  if(src.numRows() != tgt.numCols() || src.numCols() != tgt.numRows())
    throw new IllegalArgumentException("dimension do not match!");
  for(Vec v:src.vecs()) {
    if (v.isCategorical())
      throw new IllegalArgumentException("transpose can only be applied to all-numeric frames (representing a matrix)");
    if(v.length() > 1000000)
      throw new IllegalArgumentException("too many rows, transpose only works for frames with < 1M rows.");
  }
  new TransposeTsk(tgt).doAll(src);
  return tgt;
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
Val exec_check( Env env, Env.StackHelp stk, Frame tst, AST ast, Frame xfr ) {
  Val val = ast.exec(env);
  if( val.isFrame() ) {
    Frame fr = stk.track(val).getFrame();
    if( tst.numCols() != fr.numCols() || tst.numRows() != fr.numRows() )
      throw new IllegalArgumentException("ifelse test frame and other frames must match dimensions, found "+tst+" and "+fr);
    xfr.add(fr);
  }
  return val;
}
 

import water.fvec.Frame; //导入方法依赖的package包/类
public MatrixMulTsk(H2OCountedCompleter cmp, Key progressKey, Frame x, Frame y) {
  super(cmp);
  if(x.numCols() != y.numRows())
    throw new IllegalArgumentException("dimensions do not match! x.numcols = " + x.numCols() + ", y.numRows = " + y.numRows());
  _x = x;
  _y = y;
  _progressKey = progressKey;
}