static IEnumerable <double[]> GetAllImages(Image[] bigs)
{
int w = 28;
int h = 28;
int size = 2072;
for (int y = 0; y < size - h - 2; y += h)
{
for (int x = 0; x < size - w - 2; x += w)
{
Rectangle rec = new Rectangle(x, y, w + 1, h + 1);
foreach (Image img in bigs)
{
yield return(Digitalizer.GetInput((Bitmap)img, x, y, 29, 29, c => c.B > 100).ToArray());
}
}
}
}
private void button2_Click(object sender, EventArgs e)
{
//recon
if (!Loaded)
{
MessageBox.Show("Załaduj sieć neuronową!", "Warning");
}
double[] input = null;
using (Bitmap cropped = Digitalizer.AutoCrop(UserBitmap)) {
using (Bitmap sized = Digitalizer.Resize(cropped, 17, 20)) {
using (Bitmap bm = new Bitmap(29, 29)) {
Graphics gr = Graphics.FromImage(bm);
gr.Clear(Color.White);
gr.DrawImage(sized, 4, 4);
input = Digitalizer.GetInput(bm, 0, 0, bm.Width, bm.Height, x => x.Name == "ff000000").ToArray();
}
}
}
double[] output = Network.Pulse(input);
DrawReconResults(output);
}
//util
void StartLearning()
{
Stopped = false;
Thread.CurrentThread.Priority = ThreadPriority.Lowest;
Image[] bigimgs = Utility.Generate(i => Image.FromFile(@"mnist_train" + i.ToString() + ".jpg"), 10).ToArray();
int idx = 0, total = 0;
List <TrainingData> trainingList = new List <TrainingData>();
foreach (double[] input in GetAllImages(bigimgs))
{
if (Stopped)
{
return;
}
double[] output = Digitalizer.GetOutput(10, idx).ToArray();
Application.DoEvents();
trainingList.Add(new TrainingData(input, output));
GC.Collect();
GC.WaitForPendingFinalizers();
idx = (idx + 1) % 10;
if (idx == 0)
{
total++;
BPRequest request = new BPRequest(trainingList.ToArray(), 1)
{
ShuffleTrainingSet = true
};
BPResponse response = Network.BP(request);
double eps = response.Epochs[0].Epsilon;
double ms = response.BPTime.TotalMilliseconds;
trainingList.Clear();
}
}
StopLearning();
}
