public void PrepareData()
{
mnist = MnistDataSet.read_data_sets("mnist", one_hot: true);
// In this example, we limit mnist data
(Xtr, Ytr) = mnist.train.next_batch(5000); // 5000 for training (nn candidates)
(Xte, Yte) = mnist.test.next_batch(200); // 200 for testing
}
public void PrepareData()
{
mnist = MnistDataSet.read_data_sets("mnist", one_hot: true, train_size: TrainSize, validation_size: ValidationSize, test_size: TestSize);
// In this example, we limit mnist data
(Xtr, Ytr) = mnist.train.next_batch(TrainSize == null ? 5000 : TrainSize.Value / 100); // 5000 for training (nn candidates)
(Xte, Yte) = mnist.test.next_batch(TestSize == null ? 200 : TestSize.Value / 100); // 200 for testing
}
public static void RunPredictTest()
{
LenetModel model = new LenetModel();
model.Load(@"lenet.params");
MnistDataSet ds = new MnistDataSet(@"C:\素材\data\train-images.idx3-ubyte", @"C:\素材\data\train-labels.idx1-ubyte");
int W = 28;
int H = 28;
List <float> listData = ds.Data;
List <float> listLabel = ds.Label;
int dataCount = ds.Count;
using (FloatListHolder hData = listData.GetHolder())
using (FloatListHolder hLabel = listLabel.GetHolder())
{
NDArray data_array = new NDArray(new Shape((uint)dataCount, 1, (uint)W, (uint)H), Context.Gpu(),
false); // store in main memory, and copy to
// device memory while training
NDArray label_array = new NDArray(new Shape((uint)dataCount), Context.Gpu(),
false); // it's also ok if just store them all in device memory
data_array.SyncCopyFromCPU(hData.Handle, (ulong)(dataCount * W * H));
label_array.SyncCopyFromCPU(hLabel.Handle, (ulong)dataCount);
data_array.WaitToRead();
label_array.WaitToRead();
for (int i = 0; i < 100; i++)
{
NDArray data = data_array.Slice((uint)i, (uint)i + 1);
String output = model.Predict(data);
MnistDataSet.PrintImage(output, data);
System.Threading.Thread.Sleep(1000);
}
}
}
public void PrepareData()
{
mnist = MnistDataSet.read_data_sets("mnist", one_hot: true, train_size: train_size, validation_size: validation_size, test_size: test_size);
full_data_x = mnist.train.images;
// download graph meta data
string url = "https://raw.githubusercontent.com/SciSharp/TensorFlow.NET/master/graph/kmeans.meta";
Web.Download(url, "graph", "kmeans.meta");
}
public void PrepareData()
{
mnist = MnistDataSet.read_data_sets("mnist", one_hot: true, train_size: train_size, validation_size: validation_size, test_size: test_size);
}
public void PrepareData()
{
mnist = MnistDataSet.read_data_sets("mnist", one_hot: true);
full_data_x = mnist.train.images;
}
public void PrepareData()
{
mnist = MnistDataSet.read_data_sets("mnist", one_hot: true);
}
async Task TrainNetworkAsync()
{
try {
var weightsName = "mnist5.weights";
var weightsPath = Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.MyDocuments), weightsName);
var hasWeights = File.Exists(weightsPath);
bool needsTrain = !hasWeights;
needsTrain = true;
//
// Create the network
//
var network = new RecognizerNetwork();
network.ImagesShown += Network_ImagesShown;
Console.WriteLine(network);
//
// Read previously trained weights
//
if (hasWeights)
{
await network.ReadAsync(weightsPath);
}
//
// Load the data set
//
var dataSet = new MnistDataSet(seed: 42);
//
// Train the network
//
if (needsTrain)
{
await network.TrainAsync(dataSet);
}
//
// Save the network if training went well
//
if (network.WeightsAreValid())
{
if (needsTrain)
{
await network.WriteAsync(weightsPath);
}
//
// Start predicting
//
await network.PredictAsync(dataSet);
}
else
{
Console.WriteLine("Bad weights");
}
//
// All done
//
network.ImagesShown -= Network_ImagesShown;
}
catch (Exception ex) {
Console.WriteLine(ex);
}
}
public void PrepareData()
{
mnist = MnistDataSet.read_data_sets("logistic_regression", one_hot: true);
}
public void Run()
{
Symbol lenet = CreateLenet();
//Symbol lenet = CreateFrom(@"C:\Works\Projects\80_Project_Python\mxnet\ocr\model\mnist-symbol.json");
/*setup basic configs*/
int valFold = 1;
int W = 28;
int H = 28;
uint batchSize = 256;
int maxEpoch = 20;
float learning_rate = 0.05f;
float weight_decay = 0.0001f;
MnistDataSet ds = new MnistDataSet(@"C:\素材\data\train-images.idx3-ubyte", @"C:\素材\data\train-labels.idx1-ubyte");
//ds.Print();
List <float> listData = ds.Data;
List <float> listLabel = ds.Label;
int dataCount = ds.Count;
using (FloatListHolder hData = listData.GetHolder())
using (FloatListHolder hLabel = listLabel.GetHolder())
{
NDArray data_array = new NDArray(new Shape((uint)dataCount, 1, (uint)W, (uint)H), ctx_cpu,
false); // store in main memory, and copy to
// device memory while training
NDArray label_array = new NDArray(new Shape((uint)dataCount), ctx_cpu,
false); // it's also ok if just store them all in device memory
data_array.SyncCopyFromCPU(hData.Handle, (ulong)(dataCount * W * H));
label_array.SyncCopyFromCPU(hLabel.Handle, (ulong)dataCount);
data_array.WaitToRead();
label_array.WaitToRead();
uint train_num = (uint)(dataCount * (1 - valFold / 10.0));
train_data = data_array.Slice(0, train_num);
train_label = label_array.Slice(0, train_num);
val_data = data_array.Slice(train_num, (uint)dataCount);
val_label = label_array.Slice(train_num, (uint)dataCount);
Console.WriteLine("Data loaded ok!");
/*init some of the args*/
args_map["data"] = data_array.Slice(0, (uint)batchSize).Clone(ctx_dev);
args_map["data_label"] = label_array.Slice(0, (uint)batchSize).Clone(ctx_dev);
NDArray.WaitAll();
Console.WriteLine("Data sliced ok!");
lenet.InferArgsMap(ctx_dev, args_map, args_map, new XavierInitializer(2));
Optimizer opt = OptimizerRegistry.Find("sgd");
opt.SetParam("momentum", 0.9).SetParam("rescale_grad", 1.0 / batchSize);
for (int ITER = 0; ITER < maxEpoch; ++ITER)
{
Stopwatch sw = new Stopwatch();
sw.Start();
uint start_index = 0;
while (start_index < train_num)
{
if (start_index + batchSize > train_num)
{
start_index = train_num - batchSize;
}
args_map["data"] = train_data.Slice(start_index, start_index + batchSize).Clone(ctx_dev);
args_map["data_label"] = train_label.Slice(start_index, start_index + batchSize).Clone(ctx_dev);
start_index += batchSize;
NDArray.WaitAll();
Executor exe = lenet.SimpleBind(ctx_dev, args_map, new XavierInitializer(2));
exe.Forward(true);
exe.Backward();
exe.UpdateAll(opt, learning_rate, weight_decay);
exe.Dispose();
}
sw.Stop();
Console.WriteLine("Epoch[" + ITER + "] validation accuracy = " + ValAccuracy(batchSize, lenet) + ", time cost " + sw.Elapsed.TotalSeconds.ToString("0.00") + "s");
}
}
NDArray.Save("lenet.params", args_map);
}