public async Task RestoreClassifier_ReturnsValidObjectToClient()
{
var bytes = BitConverter.GetBytes(0);
Assert.That(bytes.Length, Is.EqualTo(4));
var endpoint = new Uri("tcp://localhost:9211");
var data = Functions.NormalRandomDataset(3, 10).Select(x => new
{
x = x,
y = Math.Log(x)
}).AsQueryable();
using (var blobs = new InMemoryBlobStore())
using (var server = new RemoteClassifierTrainingServer(endpoint, blobs))
using (var client = new RemoteClassifierTrainingClient(endpoint))
{
client.Timeout = 10000;
server.Start();
var trainingSet = data.CreatePipeline().AsTrainingSet(x => x.x > 10 ? 'a' : 'b');
var keyAndclassifier = await client.CreateClassifier(trainingSet, true);
var restoredClassifier = await client.RestoreClassifier(keyAndclassifier.Key, trainingSet.FeaturePipeline.FeatureExtractor, 'a');
var results = restoredClassifier.Classify(new
{
x = 7d,
y = 4.543d
});
}
}
public async Task CreateClassifier_ThenExtendClassifier_ReturnsTheSameUri()
{
var bytes = BitConverter.GetBytes(0);
Assert.That(bytes.Length, Is.EqualTo(4));
var endpoint = new Uri("tcp://localhost:9212");
var data = Functions.NormalRandomDataset(3, 10).Select(x => new
{
x = x,
y = Math.Log(x)
}).AsQueryable();
using (var blobs = new InMemoryBlobStore())
using (var server = new RemoteClassifierTrainingServer(endpoint, blobs))
using (var client = new RemoteClassifierTrainingClient(endpoint))
{
#if DEBUG
client.Timeout = 15000;
#endif
server.Start();
var trainingSet1 = data.Take(100).CreatePipeline().AsTrainingSet(x => x.x > 10 ? 'a' : 'b');
var trainingSet2 = data.Skip(100).CreatePipeline().AsTrainingSet(x => x.x > 10 ? 'a' : 'b');
var keyAndclassifier = await client.CreateClassifier(trainingSet1, true);
Assert.That(keyAndclassifier.Value, Is.Not.Null);
Console.WriteLine(keyAndclassifier.Key);
var keyAndclassifier2 = await client.ExtendClassifier(trainingSet2, keyAndclassifier.Key);
Console.WriteLine(keyAndclassifier2.Key);
Assert.That(keyAndclassifier.Key, Is.EqualTo(keyAndclassifier2.Key));
Assert.That(keyAndclassifier2.Value, Is.Not.Null);
}
}
public async Task CreateClassifier_SavesOutput()
{
var bytes = BitConverter.GetBytes(0);
Assert.That(bytes.Length, Is.EqualTo(4));
var endpoint = new Uri("tcp://localhost:9212");
var data = Functions.NormalRandomDataset(3, 10).Select(x => new
{
x = x,
y = Math.Log(x)
}).AsQueryable();
using (var blobs = new InMemoryBlobStore())
using (var server = new RemoteClassifierTrainingServer(endpoint, blobs))
using (var client = new RemoteClassifierTrainingClient(endpoint))
{
#if DEBUGGING
client.Timeout = 20000;
#endif
server.Start();
var trainingSet = data.CreatePipeline().AsTrainingSet(x => x.x > 10 ? 'a' : 'b');
var keyAndclassifier = await client.CreateClassifier(trainingSet, true);
// We should be able to restore the raw blob back into a network
var nn = new MultilayerNetwork(new NetworkParameters(1, 1));
blobs.Restore(new string(keyAndclassifier.Key.PathAndQuery.Skip(1).ToArray()), nn);
// We should receive a valid classifier object back
var cls = keyAndclassifier.Value.Classify(new
{
x = 12.432,
y = Math.Log(12.432)
});
foreach (var c in cls)
{
Console.WriteLine(c);
}
Assert.That(cls.Any());
}
}
public async Task CreateClassifier_TwoClients()
{
var bytes = BitConverter.GetBytes(0);
Assert.That(bytes.Length, Is.EqualTo(4));
var endpoint = new Uri("tcp://localhost:9214");
var data = Functions.NormalRandomDataset(3, 10).Select(x => new
{
x = x,
y = Math.Log(x)
}).AsQueryable();
using (var blobs = new InMemoryBlobStore())
using (var server = new RemoteClassifierTrainingServer(endpoint, blobs))
using (var client1 = new RemoteClassifierTrainingClient(endpoint))
using (var client2 = new RemoteClassifierTrainingClient(endpoint))
{
client1.Timeout = 5000;
client2.Timeout = 5000;
server.Start();
var trainingSet = data.CreatePipeline().AsTrainingSet(x => x.x > 10 ? 'a' : 'b');
var task1 = client1.CreateClassifier(trainingSet, true);
var task2 = client2.CreateClassifier(trainingSet, true);
await task1;
await task2;
}
}