//Static methods
/// <summary>
/// Loads the data and prepares PatternBundle.
/// 1st row of the file must start with the #RepetitiveGroupOfAttributes keyword followed by
/// attribute names.
/// 2nd row of the file must start with the #Outputs keyword followed by
/// output field names.
/// 3rd+ rows are the data rows.
/// The data row must begin with at least one complete set of values for defined repetitive attributes.
/// The data row must end with values of defined output fields.
/// </summary>
/// <param name="fileName"> Data file name </param>
/// <param name="inputFieldNameCollection"> Input fields to be extracted from a file</param>
/// <param name="outputFieldNameCollection"> Output fields to be extracted from a file</param>
public static PatternBundle LoadFromCsv(string fileName,
List <string> inputFieldNameCollection,
List <string> outputFieldNameCollection
)
{
PatternBundle bundle = new PatternBundle();
using (StreamReader streamReader = new StreamReader(new FileStream(fileName, FileMode.Open)))
{
List <int> inputFieldGrpIndexes = new List <int>();
List <int> outputFieldIndexes = new List <int>();
//The first row contains the "#RepetitiveGroupOfAttributes" keyword followed by name(s) of attribute(s)
string delimitedRepetitiveGroupOfAttributes = streamReader.ReadLine();
if (!delimitedRepetitiveGroupOfAttributes.StartsWith("#RepetitiveGroupOfAttributes"))
{
throw new FormatException("1st row of the file doesn't start with the #RepetitiveGroupOfAttributes keyword.");
}
//What data delimiter is used?
char csvDelimiter = DelimitedStringValues.RecognizeDelimiter(delimitedRepetitiveGroupOfAttributes);
//Split column names
DelimitedStringValues repetitiveGroupOfAttributes = new DelimitedStringValues(csvDelimiter);
repetitiveGroupOfAttributes.LoadFromString(delimitedRepetitiveGroupOfAttributes);
repetitiveGroupOfAttributes.RemoveTrailingWhites();
//Check if the recognized data delimiter works properly
if (repetitiveGroupOfAttributes.NumOfStringValues < 2)
{
throw new FormatException("The value delimiter was not recognized or missing repetitive attribute(s) name(s).");
}
//Remove the #RepetitiveGroupOfAttributes keyword from the collection
repetitiveGroupOfAttributes.RemoveAt(0);
//Check if attribute names match with the input fields collection
if (repetitiveGroupOfAttributes.NumOfStringValues < inputFieldNameCollection.Count)
{
throw new FormatException("Inconsistent number of attributes in the file and number of specified input fields.");
}
foreach (string inputFieldName in inputFieldNameCollection)
{
int index = repetitiveGroupOfAttributes.IndexOf(inputFieldName);
if (index < 0)
{
throw new FormatException($"Input field name {inputFieldName} was not found among the repetitive attributes specified in the file.");
}
inputFieldGrpIndexes.Add(index);
}
//The second row contains the "#Outputs" keyword followed by name(s) of output class(es) or values(s)
string delimitedOutputNames = streamReader.ReadLine();
if (!delimitedOutputNames.StartsWith("#Outputs"))
{
throw new FormatException("2nd row of the file doesn't start with the #Outputs keyword.");
}
DelimitedStringValues outputNames = new DelimitedStringValues(csvDelimiter);
outputNames.LoadFromString(delimitedOutputNames);
outputNames.RemoveTrailingWhites();
//Remove the #Outputs keyword from the collection
outputNames.RemoveAt(0);
//Check if the there is at least one output name
if (outputNames.NumOfStringValues < 1)
{
throw new FormatException("Missing output name(es).");
}
//Check if output names match with the output fields collection
if (outputNames.NumOfStringValues < outputFieldNameCollection.Count)
{
throw new FormatException("Inconsistent number of outputs in the file and number of specified output fields.");
}
foreach (string outputFieldName in outputFieldNameCollection)
{
int index = outputNames.IndexOf(outputFieldName);
if (index < 0)
{
throw new FormatException($"Output field name {outputFieldName} was not found among the outputs specified in the file.");
}
outputFieldIndexes.Add(index);
}
//Load data
DelimitedStringValues dataRow = new DelimitedStringValues(csvDelimiter);
while (!streamReader.EndOfStream)
{
dataRow.LoadFromString(streamReader.ReadLine());
dataRow.RemoveTrailingWhites();
//Check data length
if (dataRow.NumOfStringValues < repetitiveGroupOfAttributes.NumOfStringValues + outputNames.NumOfStringValues ||
((dataRow.NumOfStringValues - outputNames.NumOfStringValues) % repetitiveGroupOfAttributes.NumOfStringValues) != 0)
{
throw new FormatException("Incorrect length of data row.");
}
//Pattern data
List <double[]> patternData = new List <double[]>();
for (int grpIdx = 0; grpIdx < (dataRow.NumOfStringValues - outputNames.NumOfStringValues) / repetitiveGroupOfAttributes.NumOfStringValues; grpIdx++)
{
double[] inputVector = new double[inputFieldGrpIndexes.Count];
for (int i = 0; i < inputFieldGrpIndexes.Count; i++)
{
inputVector[i] = dataRow.GetValue(grpIdx * repetitiveGroupOfAttributes.NumOfStringValues + inputFieldGrpIndexes[i]).ParseDouble(true, $"Can't parse double data value {dataRow.GetValue(grpIdx * repetitiveGroupOfAttributes.NumOfStringValues + inputFieldGrpIndexes[i])}.");
}
patternData.Add(inputVector);
}//grpIdx
//Output data
double[] outputVector = new double[outputFieldIndexes.Count];
int dataRowStartIdx = dataRow.NumOfStringValues - outputNames.NumOfStringValues;
for (int i = 0; i < outputFieldIndexes.Count; i++)
{
outputVector[i] = dataRow.GetValue(dataRowStartIdx + outputFieldIndexes[i]).ParseDouble(true, $"Can't parse double value {dataRow.GetValue(dataRowStartIdx + outputFieldIndexes[i])}.");
}
bundle.AddPair(patternData, outputVector);
} //while !EOF
} //using streamReader
return(bundle);
} //LoadFromCsv
/// <summary>
/// Loads the data and prepares PatternBundle.
/// 1st row of the file must start with the #RepetitiveGroupOfAttributes keyword followed by
/// attribute names.
/// 2nd row of the file must start with the #Outputs keyword followed by
/// output field names.
/// 3rd+ rows are the data rows.
/// The data row must begin with at least one set of values for defined repetitive attributes.
/// The data row must end with a value for each defined output.
/// </summary>
/// <param name="classification">
/// In case of classification the standardization and reserve ratio are not applied on output fields.
/// </param>
/// <param name="fileName">
/// Data file name
/// </param>
/// <param name="inputFieldNameCollection">
/// Input fields
/// </param>
/// <param name="outputFieldNameCollection">
/// Output fields
/// </param>
/// <param name="normRange">
/// Range of normalized values
/// </param>
/// <param name="normReserveRatio">
/// Reserve held by a normalizer to cover cases where future data exceeds a known range of sample data.
/// </param>
/// <param name="dataStandardization">
/// Specifies whether to apply data standardization to input data.
/// Output data is never standardized.
/// </param>
/// <param name="bundleNormalizer">
/// Returned initialized instance of BundleNormalizer.
/// </param>
public static PatternBundle Load(bool classification,
string fileName,
List <string> inputFieldNameCollection,
List <string> outputFieldNameCollection,
Interval normRange,
double normReserveRatio,
bool dataStandardization,
out BundleNormalizer bundleNormalizer
)
{
PatternBundle bundle = new PatternBundle();
bundleNormalizer = new BundleNormalizer(normRange, normReserveRatio, dataStandardization, classification ? 0 : normReserveRatio, classification ? false : dataStandardization);
using (StreamReader streamReader = new StreamReader(new FileStream(fileName, FileMode.Open)))
{
//The first row contains the "#RepetitiveGroupOfAttributes" keyword followed by name(s) of attribute(s)
string delimitedRepetitiveGroupOfAttributes = streamReader.ReadLine();
if (!delimitedRepetitiveGroupOfAttributes.StartsWith("#RepetitiveGroupOfAttributes"))
{
throw new FormatException("1st row of the file doesn't start with the #RepetitiveGroupOfAttributes keyword.");
}
//What data delimiter is used?
char csvDelimiter = DelimitedStringValues.RecognizeDelimiter(delimitedRepetitiveGroupOfAttributes);
//Split column names
DelimitedStringValues repetitiveGroupOfAttributes = new DelimitedStringValues(csvDelimiter);
repetitiveGroupOfAttributes.LoadFromString(delimitedRepetitiveGroupOfAttributes);
repetitiveGroupOfAttributes.RemoveTrailingWhites();
//Check if the recognized data delimiter works properly
if (repetitiveGroupOfAttributes.NumOfStringValues < 2)
{
throw new FormatException("The value delimiter was not recognized or missing repetitive attribute(s) name(s).");
}
//Remove the #RepetitiveGroupOfAttributes keyword from the collection
repetitiveGroupOfAttributes.RemoveAt(0);
//Check if attribute names match with the input fields collection
if (repetitiveGroupOfAttributes.NumOfStringValues != inputFieldNameCollection.Count)
{
throw new FormatException("Different number of attributes in the file and number of specified input fields.");
}
foreach (string inputFieldName in inputFieldNameCollection)
{
if (repetitiveGroupOfAttributes.IndexOf(inputFieldName) < 0)
{
throw new FormatException($"Input field name {inputFieldName} was not found among the repetitive attributes specified in the file.");
}
}
//The second row contains the "#Outputs" keyword followed by name(s) of output class(es) or values(s)
string delimitedOutputNames = streamReader.ReadLine();
if (!delimitedOutputNames.StartsWith("#Outputs"))
{
throw new FormatException("2nd row of the file doesn't start with the #Outputs keyword.");
}
DelimitedStringValues outputNames = new DelimitedStringValues(csvDelimiter);
outputNames.LoadFromString(delimitedOutputNames);
outputNames.RemoveTrailingWhites();
//Check if the there is at least one output name
if (outputNames.NumOfStringValues < 2)
{
throw new FormatException("Missing output name(es).");
}
//Remove the #Outputs keyword from the collection
outputNames.RemoveAt(0);
//Check if output names match with the output fields collection
if (outputNames.NumOfStringValues != outputFieldNameCollection.Count)
{
throw new FormatException("Different number of outputs in the file and number of specified output fields.");
}
foreach (string outputFieldName in outputFieldNameCollection)
{
if (outputNames.IndexOf(outputFieldName) < 0)
{
throw new FormatException($"Output field name {outputFieldName} was not found among the outputs specified in the file.");
}
}
//Bundle handler setup
foreach (string attrName in repetitiveGroupOfAttributes.StringValueCollection)
{
bundleNormalizer.DefineField(attrName, attrName);
bundleNormalizer.DefineInputField(attrName);
}
foreach (string outputName in outputNames.StringValueCollection)
{
bundleNormalizer.DefineField(outputName, outputName);
bundleNormalizer.DefineOutputField(outputName);
}
bundleNormalizer.FinalizeStructure();
//Load data
DelimitedStringValues dataRow = new DelimitedStringValues(csvDelimiter);
while (!streamReader.EndOfStream)
{
dataRow.LoadFromString(streamReader.ReadLine());
dataRow.RemoveTrailingWhites();
//Check data length
if (dataRow.NumOfStringValues < repetitiveGroupOfAttributes.NumOfStringValues + outputNames.NumOfStringValues ||
((dataRow.NumOfStringValues - outputNames.NumOfStringValues) % repetitiveGroupOfAttributes.NumOfStringValues) != 0)
{
throw new FormatException("Incorrect length of data row.");
}
//Pattern data
List <double[]> patternData = new List <double[]>();
for (int grpIdx = 0; grpIdx < (dataRow.NumOfStringValues - outputNames.NumOfStringValues) / repetitiveGroupOfAttributes.NumOfStringValues; grpIdx++)
{
double[] inputVector = new double[repetitiveGroupOfAttributes.NumOfStringValues];
for (int attrIdx = 0; attrIdx < repetitiveGroupOfAttributes.NumOfStringValues; attrIdx++)
{
inputVector[attrIdx] = dataRow.GetValue(grpIdx * repetitiveGroupOfAttributes.NumOfStringValues + attrIdx).ParseDouble(true, "Can't parse double data value.");
} //attrIdx
patternData.Add(inputVector);
} //grpIdx
//Output data
double[] outputVector = new double[outputNames.NumOfStringValues];
for (int outputIdx = (dataRow.NumOfStringValues - outputNames.NumOfStringValues), i = 0; outputIdx < dataRow.NumOfStringValues; outputIdx++, i++)
{
outputVector[i] = dataRow.GetValue(outputIdx).ParseDouble(true, $"Can't parse double value {dataRow.GetValue(outputIdx)}.");
} //outputIdx
bundle.AddPair(patternData, outputVector);
} //while !EOF
} //using streamReader
//Data normalization
bundleNormalizer.Normalize(bundle);
return(bundle);
} //Load
/// <summary>
/// Loads the data and prepares PredictionBundle.
/// The first line of the csv file must be field names. These field names must
/// match the names of the input and output fields.
/// </summary>
/// <param name="fileName">
/// Data file name
/// </param>
/// <param name="inputFieldNameCollection">
/// Input fields
/// </param>
/// <param name="outputFieldNameCollection">
/// Output fields
/// </param>
/// <param name="normRange">
/// Range of normalized values
/// </param>
/// <param name="normReserveRatio">
/// Reserve held by a normalizer to cover cases where future data exceeds a known range of sample data.
/// </param>
/// <param name="dataStandardization">
/// Specifies whether to apply data standardization
/// </param>
/// <param name="singleNormalizer">
/// Use true if all input and output fields are about the same range of values.
/// </param>
/// <param name="bundleNormalizer">
/// Returned initialized instance of BundleNormalizer.
/// </param>
/// <param name="remainingInputVector">
/// Returned the last input vector unused in the bundle.
/// </param>
public static TimeSeriesBundle Load(string fileName,
List <string> inputFieldNameCollection,
List <string> outputFieldNameCollection,
Interval normRange,
double normReserveRatio,
bool dataStandardization,
bool singleNormalizer,
out BundleNormalizer bundleNormalizer,
out double[] remainingInputVector
)
{
TimeSeriesBundle bundle = null;
bundleNormalizer = new BundleNormalizer(normRange, normReserveRatio, dataStandardization, normReserveRatio, dataStandardization);
using (StreamReader streamReader = new StreamReader(new FileStream(fileName, FileMode.Open)))
{
List <int> fieldIndexes = new List <int>();
List <double[]> allData = new List <double[]>();
//First row contains column names (data fields)
string delimitedColumnNames = streamReader.ReadLine();
//What data delimiter is used?
char csvDelimiter = DelimitedStringValues.RecognizeDelimiter(delimitedColumnNames);
//Split column names
DelimitedStringValues columnNames = new DelimitedStringValues(csvDelimiter);
columnNames.LoadFromString(delimitedColumnNames);
//Check if the recognized data delimiter works properly
if (columnNames.NumOfStringValues < inputFieldNameCollection.Count)
{
throw new FormatException("1st row of the file doesn't contain delimited column names or the value delimiter was not properly recognized.");
}
//Define fields
foreach (string name in inputFieldNameCollection)
{
if (!bundleNormalizer.IsFieldDefined(name))
{
bundleNormalizer.DefineField(name, singleNormalizer ? "COMMON" : name);
fieldIndexes.Add(columnNames.IndexOf(name));
}
bundleNormalizer.DefineInputField(name);
}
foreach (string name in outputFieldNameCollection)
{
if (!bundleNormalizer.IsFieldDefined(name))
{
bundleNormalizer.DefineField(name, singleNormalizer ? "COMMON" : name);
fieldIndexes.Add(columnNames.IndexOf(name));
}
bundleNormalizer.DefineOutputField(name);
}
//Finalize structure
bundleNormalizer.FinalizeStructure();
//Load all relevant data
DelimitedStringValues dataRow = new DelimitedStringValues(csvDelimiter);
while (!streamReader.EndOfStream)
{
dataRow.LoadFromString(streamReader.ReadLine());
double[] vector = new double[fieldIndexes.Count];
for (int i = 0; i < fieldIndexes.Count; i++)
{
vector[i] = dataRow.GetValue(fieldIndexes[i]).ParseDouble(true, $"Can't parse double value {dataRow.GetValue(fieldIndexes[i])}.");
}
allData.Add(vector);
}
//Create data bundle
remainingInputVector = bundleNormalizer.CreateBundleFromVectorCollection(allData, true, out bundle);
}
return(bundle);
} //Load