public IActionResult PredictCharacter(string data)
{
OCRCharacter charGuess = createCharacterFromBytes(data);
int guess = MulticlassClassifier.predictChar(charGuess, ML_TRAINED_MODEL);
return(Json(guess.ToString()));
}
private static string runNearesNeighbour(IDistance distCalc, string predictTest)
{
List <OCRCharacter> charactersTrain = new List <OCRCharacter>();
OCRCharacter charGuess = createCharacterFromBytes(predictTest);
///List<OCRCharacter> charactersTest = new List<OCRCharacter>();
///addCharacter(charactersTest, predictTest, false);
try
{
// Load the training file points
loadDataFromFile(charactersTrain, DATASET_FILE_1, DATASET_FILE_CONTENT_1);
// Load the training file points from dataset 2 as well.
loadDataFromFile(charactersTrain, DATASET_FILE_2, DATASET_FILE_CONTENT_2);
// Predict the number
int answer = NearestNeighbourClassifier.processNNAndPredict(charactersTrain, charGuess, distCalc);
return(answer.ToString());
}
catch (Exception e)
{
return("ERROR " + e.Message + " FILE: " + DATASET_FILE_1);
}
}
private void TestSlice(string[] slice, string expectation)
{
var character = new OCRCharacter(slice);
string result = OCRCharInterpreter.InterpreteChar(character);
Assert.AreEqual(expectation, result);
}
public IActionResult PredictCharacterMLP(string data)
{
if (mlp == null)
{
runMLP(-1);
}
OCRCharacter charGuess = createCharacterFromBytes(data);
int guess = mlp.recallNetworkGuess(charGuess);
return(Json(guess.ToString()));
}
public IActionResult PredictCharacterKNN(string data)
{
List <OCRCharacter> charactersTrain = new List <OCRCharacter>();
List <OCRCharacter> charactersTest = new List <OCRCharacter>();
// Load the training file points
loadDataFromFile(charactersTrain, DATASET_FILE_1, DATASET_FILE_CONTENT_1);
// Load the training file points from dataset 2 as well.
loadDataFromFile(charactersTrain, DATASET_FILE_2, DATASET_FILE_CONTENT_2);
// Load the training file points from dataset 2 as well.
///loadDataFromFile(charactersTest, DATASET_FILE_2, DATASET_FILE_CONTENT_2);
bool useDistanceScoring = true;
OCRCharacter charGuess = createCharacterFromBytes(data);
int guess = KNNClassifier.processKNNAndPredict(charactersTrain, charGuess, new EuclideanDistance(), 4, useDistanceScoring);
return(Json(guess.ToString()));
}
public static int predictChar(OCRCharacter charTest, string fileMLTrainedModel)
{
ITransformer trainedModel = mlContext.Model.Load(fileMLTrainedModel, out var modelInputSchema);
// Create prediction engine related to the loaded trained model
var predEngine = mlContext.Model.CreatePredictionEngine <InputData, OutPutData>(trainedModel);
var resultprediction1 = predEngine.Predict(new InputData()
{
Number = 1, PixelValues = charTest.getPointsFloatArr()
});
//Console.WriteLine($"Actual: 1 Predicted probability: zero: {resultprediction1.Score[0]:0.####}");
//Console.WriteLine($" One : {resultprediction1.Score[1]:0.####}");
//Console.WriteLine($" two: {resultprediction1.Score[2]:0.####}");
//Console.WriteLine($" three: {resultprediction1.Score[3]:0.####}");
//Console.WriteLine($" four: {resultprediction1.Score[4]:0.####}");
//Console.WriteLine($" five: {resultprediction1.Score[5]:0.####}");
//Console.WriteLine($" six: {resultprediction1.Score[6]:0.####}");
//Console.WriteLine($" seven: {resultprediction1.Score[7]:0.####}");
//Console.WriteLine($" eight: {resultprediction1.Score[8]:0.####}");
//Console.WriteLine($" nine: {resultprediction1.Score[9]:0.####}");
int ans = -1;
float ansVal = 0;
// Chose the one cahracter with the highest likelihood score
for (int i = 0; i < 10; i++)
{
if (resultprediction1.Score[i] > ansVal)
{
ansVal = resultprediction1.Score[i];
ans = i;
}
}
return(ans);
}
