public override float OutputsDense(Floatarray result, Floatarray x_raw)
{
CHECK_ARG(x_raw.Length() == w1.Dim(1), "x_raw.Length() == w1.Dim(1)");
Floatarray z = new Floatarray();
int sparse = PGeti("sparse");
Floatarray y = new Floatarray();
Floatarray x = new Floatarray();
x.Copy(x_raw);
mvmul0(y, w1, x);
y += b1;
for (int i = 0; i < y.Length(); i++)
{
y[i] = sigmoid(y[i]);
}
if (sparse > 0)
{
ClassifierUtil.Sparsify(y, sparse);
}
mvmul0(z, w2, y);
z += b2;
for (int i = 0; i < z.Length(); i++)
{
z[i] = sigmoid(z[i]);
}
result.Copy(z);
//int idx = NarrayUtil.ArgMax(result);
//float val = NarrayUtil.Max(result);
return(Convert.ToSingle(Math.Abs(NarrayUtil.Sum(z) - 1.0)));
}
public override void TrainDense(IDataset ds)
{
int nclasses = ds.nClasses();
float miters = PGetf("miters");
int niters = (int)(ds.nSamples() * miters);
niters = Math.Max(1000, Math.Min(10000000, niters));
double err = 0.0;
Floatarray x = new Floatarray();
Floatarray z = new Floatarray();
Floatarray target = new Floatarray(nclasses);
int count = 0;
for (int i = 0; i < niters; i++)
{
int row = i % ds.nSamples();
ds.Output(target, row);
ds.Input1d(x, row);
TrainOne(z, target, x, PGetf("eta"));
err += NarrayUtil.Dist2Squared(z, target);
count++;
}
err /= count;
Global.Debugf("info", " {4} n {0} niters={1} eta={2:0.#####} errors={3:0.########}",
ds.nSamples(), niters, PGetf("eta"), err, FullName);
}
/// <summary>
/// Pick an array element with probability proportional to exp(-cost).
/// </summary>
public static int sample_by_costs(Floatarray costs)
{
Doublearray p = new Doublearray();
p.Copy(costs);
double mincost = NarrayUtil.Min(costs);
p -= mincost;
for (int i = 0; i < p.Length(); i++)
{
p.UnsafePut1d(i, Math.Exp(-p.UnsafeAt1d(i)));
}
double sump = NarrayUtil.Sum(p);
p /= sump;
double choice = rnd.NextDouble();
double s = 0;
for (int i = 0; i < p.Length(); i++)
{
s += p[i];
if (choice < s)
{
return(i);
}
}
// shouldn't happen...
return(costs.Length() - 1);
}
public static void remove_small_components(Intarray segmentation, int w = 5, int h = 4)
{
if (NarrayUtil.Max(segmentation) > 100000)
{
throw new Exception("remove_small_components: to many segments");
}
Narray <Rect> bboxes = new Narray <Rect>();
ImgLabels.bounding_boxes(ref bboxes, segmentation);
for (int i = 1; i < bboxes.Length(); i++)
{
Rect b = bboxes[i];
if (b.Width() < w && b.Height() < h)
{
for (int x = b.x0; x < b.x1; x++)
{
for (int y = b.y0; y < b.y1; y++)
{
if (segmentation[x, y] == i)
{
segmentation[x, y] = 0;
}
}
}
}
}
}
public override void FindBestCuts()
{
unchecked
{
for (int i = 0; i < cutcosts.Length(); i++)
{
NarrayUtil.ExtPut(dimage, i, (int)(cutcosts[i] + 10), 0xff0000);
}
for (int i = 0; i < cutcosts.Length(); i++)
{
NarrayUtil.ExtPut(dimage, i, (int)(min_thresh + 10), 0x800000);
}
}
Floatarray temp = new Floatarray();
Gauss.Gauss1d(temp, cutcosts, 3.0f);
cutcosts.Move(temp);
SegmRoutine.local_minima(ref bestcuts, cutcosts, min_range, min_thresh);
for (int i = 0; i < bestcuts.Length(); i++)
{
Narray <Point> cut = cuts[bestcuts[i]];
for (int j = 0; j < cut.Length(); j++)
{
Point p = cut[j];
NarrayUtil.ExtPut(dimage, p.X, p.Y, 0x00ff00);
}
}
///-if(debug.Length > 0) write_image_packed(debug, dimage);
// dshow1d(cutcosts,"Y");
//dshow(dimage,"Y");
}
public override void SetSegmentationAndGt(Intarray segmentation, Intarray cseg, ref string text)
{
// first, set the segmentation as usual
SetSegmentation(segmentation);
// Maybe fix up the transcript (remove spaces).
gttranscript = text;
string s = text;
fixup_transcript(ref s, false);
int max_cseg = NarrayUtil.Max(cseg);
bool old_csegs = (s.Length != max_cseg);
fixup_transcript(ref gttranscript, old_csegs);
// Complain if it doesn't match.
if (gttranscript.Length != max_cseg)
{
Logger.Default.Format("transcript = '{0}'\n", gttranscript);
throw new Exception(String.Format("transcript doesn't agree with cseg (transcript {0}, cseg {1})",
gttranscript.Length, max_cseg));
}
// Now compute the correspondences between the character segmentation
// and the raw segmentation.
GrouperRoutine.segmentation_correspondences(gtsegments, segmentation, cseg);
}
public static void threshold_frac(Bytearray thresholded, Floatarray input, float frac)
{
float minofinput = NarrayUtil.Min(input);
float theta = frac * (NarrayUtil.Max(input) - minofinput) + minofinput;
binarize_with_threshold(thresholded, input, theta);
}
public void Get(Intarray r_vertices1,
Intarray r_vertices2,
Intarray r_inputs,
Intarray r_outputs,
Floatarray r_costs,
int id)
{
Intarray t_v1 = new Intarray(); // vertices
Intarray t_v2 = new Intarray(); // vertices
Intarray t_i = new Intarray(); // inputs
Intarray t_o = new Intarray(); // outputs
Floatarray t_c = new Floatarray(); // costs
int current = id;
while (current != -1)
{
t_v1.Push(v1[current]);
t_v2.Push(v2[current]);
t_i.Push(inputs[current]);
t_o.Push(outputs[current]);
t_c.Push(costs[current]);
current = parents[current];
}
NarrayUtil.Reverse(r_vertices1, t_v1);
NarrayUtil.Reverse(r_vertices2, t_v2);
NarrayUtil.Reverse(r_inputs, t_i);
NarrayUtil.Reverse(r_outputs, t_o);
NarrayUtil.Reverse(r_costs, t_c);
}
private void ProcessSegmentationMethod(object sender, RoutedEventArgs e)
{
string strSermenterName = (sender as Control).Tag.ToString();
ISegmentLine segmenter = ComponentCreator.MakeComponent <ISegmentLine>(strSermenterName);
if (segmenter == null || currBookLine == null)
{
return;
}
// приведем к чернобелому
Bytearray image = currBookLine.ImageBytearray;
//IBinarize binarizer = new BinarizeByOtsu();
//binarizer.Binarize(image, image);
OcrRoutine.binarize_simple(image, image);
// сегментация
Intarray charseg = new Intarray();
segmenter.Charseg(ref charseg, image);
// фон равен 0
SegmRoutine.make_line_segmentation_black(charseg);
// удалим маленькие сегменты
SegmRoutine.remove_small_components(charseg, 3, 3);
ImgLabels.renumber_labels(charseg, 1);
currBookLine.CharsegIntarray = charseg;
CurrSegmentsCount = NarrayUtil.Max(charseg);
// Show segmented image
ShowCharsegImage(charseg,
(currBookLine.HaveTranscript && CurrSegmentsCount == currBookLine.Transcript.Length) ?
currBookLine.Transcript : "");
// to enable save button
EnableCharsegCmdButtons();
}
public static void rseg_to_cseg(Intarray cseg, Intarray rseg, Intarray ids)
{
Intarray map = new Intarray(NarrayUtil.Max(rseg) + 1);
map.Fill(0);
int color = 0;
for (int i = 0; i < ids.Length(); i++)
{
if (ids[i] == 0)
{
continue;
}
color++;
int start = ids[i] >> 16;
int end = ids[i] & 0xFFFF;
if (start > end)
{
throw new Exception("segmentation encoded in IDs looks seriously broken!");
}
if (start >= map.Length() || end >= map.Length())
{
throw new Exception("segmentation encoded in IDs doesn't fit!");
}
for (int j = start; j <= end; j++)
{
map[j] = color;
}
}
cseg.MakeLike(rseg);
for (int i = 0; i < cseg.Length1d(); i++)
{
cseg.Put1d(i, map[rseg.At1d(i)]);
}
}
/// <summary>
/// Merge segments from start to end.
/// </summary>
/// <param name="cseg">Output</param>
/// <param name="rseg">Input</param>
/// <param name="start">start merge position</param>
/// <param name="end">end merge position</param>
public static void rseg_to_cseg(Intarray cseg, Intarray rseg, int start, int end)
{
int maxSegNum = NarrayUtil.Max(rseg);
if (start > end)
{
throw new Exception("segmentation encoded in IDs looks seriously broken!");
}
if (start > maxSegNum || end > maxSegNum)
{
throw new Exception("segmentation encoded in IDs doesn't fit!");
}
Intarray map = new Intarray(maxSegNum + 1);
map.Fill(0);
int color = 1;
for (int i = 1; i <= maxSegNum; i++)
{
map[i] = color;
if (i < start || i >= end)
{
color++;
}
}
cseg.MakeLike(rseg);
for (int i = 0; i < cseg.Length1d(); i++)
{
cseg.Put1d(i, map[rseg.At1d(i)]);
}
}
/// <summary>
/// Return the segmentation-derived mask for the character.
/// This may optionally be grown by some pixels.
/// </summary>
public override void GetMask(out Rect r, ref Bytearray outmask, int index, int grow)
{
r = boxes.At1d(index).Grow(grow);
r.Intersect(new Rect(0, 0, labels.Dim(0), labels.Dim(1)));
if (fullheight)
{
r.y0 = 0;
r.y1 = labels.Dim(1);
}
int x = r.x0, y = r.y0, w = r.Width(), h = r.Height();
Intarray segs = segments.At1d(index);
outmask.Resize(w, h);
outmask.Fill(0);
for (int i = 0; i < w; i++)
{
for (int j = 0; j < h; j++)
{
int label = labels[x + i, y + j];
if (NarrayUtil.first_index_of(segs, label) >= 0)
{
outmask[i, j] = (byte)255;
}
}
}
if (grow > 0)
{
Morph.binary_dilate_circle(outmask, grow);
}
}
public void InitData(IDataset ds, int nhidden, Intarray newc2i = null, Intarray newi2c = null)
{
CHECK_ARG(nhidden > 1 && nhidden < 1000000, "nhidden > 1 && nhidden < 1000000");
int ninput = ds.nFeatures();
int noutput = ds.nClasses();
w1.Resize(nhidden, ninput);
b1.Resize(nhidden);
w2.Resize(noutput, nhidden);
b2.Resize(noutput);
Intarray indexes = new Intarray();
NarrayUtil.RPermutation(indexes, ds.nSamples());
Floatarray v = new Floatarray();
for (int i = 0; i < w1.Dim(0); i++)
{
int row = indexes[i];
ds.Input1d(v, row);
float normv = (float)NarrayUtil.Norm2(v);
v /= normv * normv;
NarrayRowUtil.RowPut(w1, i, v);
}
ClassifierUtil.fill_random(b1, -1e-6f, 1e-6f);
ClassifierUtil.fill_random(w2, -1.0f / nhidden, 1.0f / nhidden);
ClassifierUtil.fill_random(b2, -1e-6f, 1e-6f);
if (newc2i != null)
{
c2i.Copy(newc2i);
}
if (newi2c != null)
{
i2c.Copy(newi2c);
}
}
protected void Recompute()
{
nc = NarrayUtil.Max(classes) + 1;
nf = data[0].Dim(0);
CHECK_ARG(DataMin(data) >= 0 && DataMax(data) < 256, "min(data) >= 0 && max(data) < 256");
CHECK_ARG(NarrayUtil.Min(classes) >= -1 && NarrayUtil.Max(classes) < 10000,
"min(classes)>=-1 && max(classes)<10000");
CHECK_ARG(nc > 0, "nc > 0");
CHECK_ARG(nf > 0, "nf > 0");
}
protected void Recompute()
{
nc = NarrayUtil.Max(classes) + 1;
nf = data.Dim(1);
CHECK_ARG(Min(data) > -100 && Max(data) < 100, "min(data) > -100 && max(data) < 100");
CHECK_ARG(NarrayUtil.Min(classes) >= -1 && NarrayUtil.Max(classes) < 10000,
"min(classes)>=-1 && max(classes)<10000");
CHECK_ARG(nc > 0, "nc > 0");
CHECK_ARG(nf > 0, "nf > 0");
}
public override void TrainDense(IDataset ds)
{
//PSet("%nsamples", ds.nSamples());
float split = PGetf("cv_split");
int mlp_cv_max = PGeti("cv_max");
if (crossvalidate)
{
// perform a split for cross-validation, making sure
// that we don't have the same sample in both the
// test and the training set (even if the data set
// is the result of resampling)
Intarray test_ids = new Intarray();
Intarray ids = new Intarray();
for (int i = 0; i < ds.nSamples(); i++)
{
ids.Push(ds.Id(i));
}
NarrayUtil.Uniq(ids);
Global.Debugf("cvdetail", "reduced {0} ids to {1} ids", ds.nSamples(), ids.Length());
NarrayUtil.Shuffle(ids);
int nids = (int)((1.0 - split) * ids.Length());
nids = Math.Min(nids, mlp_cv_max);
for (int i = 0; i < nids; i++)
{
test_ids.Push(ids[i]);
}
NarrayUtil.Quicksort(test_ids);
Intarray training = new Intarray();
Intarray testing = new Intarray();
for (int i = 0; i < ds.nSamples(); i++)
{
int id = ds.Id(i);
if (ClassifierUtil.Bincontains(test_ids, id))
{
testing.Push(i);
}
else
{
training.Push(i);
}
}
Global.Debugf("cvdetail", "#training {0} #testing {1}",
training.Length(), testing.Length());
PSet("%ntraining", training.Length());
PSet("%ntesting", testing.Length());
Datasubset trs = new Datasubset(ds, training);
Datasubset tss = new Datasubset(ds, testing);
TrainBatch(trs, tss);
}
else
{
TrainBatch(ds, ds);
}
}
public static int binarize_simple(Bytearray result, Bytearray image)
{
int threshold = (NarrayUtil.Max(image) /* + NarrayUtil.Min(image)*/) / 2;
result.MakeLike(image);
for (int i = 0; i < image.Length1d(); i++)
{
result.Put1d(i, image.At1d(i) < threshold ? (byte)0 : (byte)255);
}
return(threshold);
}
public static void combine_segmentations(ref Intarray dst, Intarray src)
{
dst.SameDims(src);
int n = NarrayUtil.Max(dst) + 1;
for (int i = 0; i < dst.Length1d(); i++)
{
dst.Put1d(i, (dst.At1d(i) + src.At1d(i) * n));
}
ImgLabels.renumber_labels(dst, 1);
}
/// <summary>
/// Изменился текущий BookLine
/// </summary>
private void BookLine_CurrentChanged(object sender, EventArgs e)
{
if (listingDataView.View.CurrentItem == null)
{
return;
}
// сбросим кэш предыдущего елемента
if (currBookLine != null)
{
currBookLine.Image = null;
currBookLine.ImageBytearray = null;
currBookLine.CharsegIntarray = null;
}
// текущий елемент
BookLine bline = listingDataView.View.CurrentItem as BookLine;
currBookLine = bline;
// сбросим некоторые признаки
isCurrDeletedSegment = false;
// показать Charseg
if (bline != null && bline.HaveCharseg)
{
// загрузим с файла
Intarray charseg = bline.CharsegIntarray;
// занесем в кэш
bline.CharsegIntarray = charseg;
// номер максимального сегмента
CurrSegmentsCount = NarrayUtil.Max(charseg);
// отобразим картинку
ShowCharsegImage(charseg, bline.Transcript);
// активируем кнопки действий
EnableCharsegCmdButtons();
}
else
{
imgCharSeg.Source = null;
DisableCharsegCmdButtons();
}
// показать Transcript
if (bline != null && bline.HaveTranscript)
{
ShowTranscript(bline.Transcript);
}
else
{
ShowTranscript("");
}
// сбросим к началу номер начальног осегмента
numUpDnStart.Value = 1;
numUpDnEnd.Value = 2;
}
private void DoTestRecognize(LenetClassifier classifier)
{
OutputVector ov = new OutputVector();
Floatarray v = new Floatarray();
Bytearray ba = new Bytearray(1, 1);
ImgIo.read_image_gray(ba, testPngFileName);
NarrayUtil.Sub(255, ba);
v.Copy(ba);
v /= 255.0;
classifier.XOutputs(ov, v);
Console.WriteLine("Featured output class '{0}', score '{1}'", (char)ov.Key(ov.BestIndex), ov.Value(ov.BestIndex));
}
public static void binary_or(Bytearray image, Bytearray image2, int dx, int dy)
{
int w = image.Dim(0);
int h = image.Dim(1);
for (int i = 0; i < w; i++)
{
for (int j = 0; j < h; j++)
{
image[i, j] = Math.Max(image[i, j], NarrayUtil.Ext(image2, i - dx, j - dy));
}
}
}
public static void skeletal_features(Bytearray endpoints, Bytearray junctions,
Bytearray image, float presmooth, float skelsmooth)
{
Bytearray temp = new Bytearray();
temp.Copy(image);
NarrayUtil.Greater(temp, (byte)128, (byte)0, (byte)255);
if (presmooth > 0f)
{
Gauss.Gauss2d(temp, presmooth, presmooth);
NarrayUtil.Greater(temp, (byte)128, (byte)0, (byte)255);
}
}
public static double beam_search(out string result, Intarray inputs, Floatarray costs, OcroFST fst1, OcroFST fst2,
int beam_width)
{
Intarray v1 = new Intarray();
Intarray v2 = new Intarray();
Intarray o = new Intarray();
//fprintf(stderr,"starting beam search\n");
beam_search(v1, v2, inputs, o, costs, fst1, fst2, beam_width);
//fprintf(stderr,"finished beam search\n");
FstUtil.remove_epsilons(out result, o);
return(NarrayUtil.Sum(costs));
}
/// <summary>
/// Get spacing to the next component.
/// </summary>
public override int PixelSpace(int i)
{
int end = NarrayUtil.Max(segments.At1d(i));
if (end >= rboxes.Length() - 1)
{
return(-1);
}
int x0 = rboxes.At1d(end).x1;
int x1 = rboxes.At1d(end + 1).x0;
return(x1 - x0);
}
public static void binsmooth(Bytearray binary, Floatarray input, float sigma)
{
Floatarray smoothed = new Floatarray();
smoothed.Copy(input);
smoothed -= NarrayUtil.Min(smoothed);
smoothed /= NarrayUtil.Max(smoothed);
if (sigma > 0)
{
Gauss.Gauss2d(smoothed, sigma, sigma);
}
binarize_with_threshold(binary, smoothed, 0.5f);
}
public static Bitmap read_image_binary(Bytearray image, string path)
{
Bitmap bitmap = LoadBitmapFromFile(path);
image.Resize(bitmap.Width, bitmap.Height);
ImgRoutine.NarrayFromBitmap(image, bitmap);
double threshold = (NarrayUtil.Min(image) + NarrayUtil.Max(image)) / 2.0;
for (int i = 0; i < image.Length1d(); i++)
{
image.Put1d(i, (byte)((image.At1d(i) < threshold) ? 0 : 255));
}
return(bitmap);
}
public override void Add(Floatarray v, int c)
{
CHECK_ARG(NarrayUtil.Min(v) > -1.2f && NarrayUtil.Max(v) < 1.2f, "float8: value out of range (-1.2..1.2)");
CHECK_ARG(c >= -1, "c>=-1");
if (c >= nc)
{
nc = c + 1;
}
if (nf < 0)
{
nf = v.Length();
}
RowPush(data, v);
classes.Push(c);
}
public static double Perplexity(Floatarray weights)
{
Floatarray w = new Floatarray();
w.Copy(weights);
w /= NarrayUtil.Sum(w);
double total = 0.0;
for (int i = 0; i < w.Length(); i++)
{
float value = w[i];
total += value * Math.Log(value);
}
return(Math.Exp(-total));
}
/// <summary>
/// Randomly permute it so that we train in random order
/// </summary>
public void Shuffle()
{
bool randomize = PGetb("randomize");
if (randomize)
{
Intarray permutation = new Intarray(all_lines.Dim(0));
for (int i = 0; i < all_lines.Dim(0); i++)
{
permutation[i] = i;
}
NarrayUtil.RandomlyPermute(permutation);
NarrayRowUtil.RowPermute(all_lines, permutation);
}
}
public static double a_star(out string result, OcroFST fst)
{
result = "";
Intarray inputs = new Intarray();
Intarray vertices = new Intarray();
Intarray outputs = new Intarray();
Floatarray costs = new Floatarray();
if (!a_star(inputs, vertices, outputs, costs, fst))
{
return(1e38);
}
FstUtil.remove_epsilons(out result, outputs);
return(NarrayUtil.Sum(costs));
}
