www.deepsignal.tech
The following section describes the different model parameters that are in the DSTCreateModelBase and DSTPredictModelBase ninjascript files. Many parameters can either be set under the Training Parameters section in Strategy Analyzer or can be hard coded in your strategy.
Parameter Parameter Name Description
Version |
Deep Signal™ Library Version |
Version of the Deep Signal Library |
DSTLibraryPath |
Deep Signal™ Library File Path |
File location of Deep Signal™ Library |
BaseFolder |
Base Folder Name |
Deep Signal™ Data folder to be created to save data |
LogMessages |
Log Messages |
Messages can be sent to the Ninjascript Output Window, to the log file or both |
BarsToTarget |
Bars To Target |
The number of bars needed to reach a long or short sell profit target |
TicksUpToTarget |
Ticks Up To Target |
The number of ticks up in x bars that the price must move to qualify our long profit target |
TicksDownToTarget |
Ticks Down To Target |
The number of ticks down in x bars that the price must move to qualify our short profit target |
UseStop |
Use Stop |
If Use Stop is True then our model will not use a buy/sell signal that gets stopped out. If Use Stop is False then our model will include signals that hit the profit target even if the instrument got stopped out. |
StopTicks |
Stop (Ticks) |
The stop price in ticks from the trade entry. If the price moves against the trade by x ticks then the algorithm assumes the strategy has stopped out. |
WindowSizeBars |
Pre Signal Window Size (1-50) |
The number of bars to include for data analysis up the point where we see a profit target hit. For example if Window Size is 10 then 10 bars of data before we hit the profit target will be included in our analysis. |
UseCloseOrOpenForPrediction |
Use Close or Open |
When we do our prediction either we try to predict either the close prices or open prices |
DontAllowPredictedPricesBetweenDays |
Don't Predict Next Day |
If we are trying to look 10 bars into the future to see if we go up 10 ticks then don't use data if the 10 bars goes into the opening of the next day. |
BinaryClassificationMetricStr |
Binary Classification Metric |
The following are classification metrics provided by Microsoft that are passed to the Microsoft ML.Net Library when training a model. When training, the algorithm will try to optimize for the selected metric. The default is the F1 Score.
Remarks: The area under the precision/recall curve is a single number summary of the information in the precision/recall curve. It is increasingly used in the machine learning community, particularly for imbalanced data sets where one class is observed more frequently than the other. On these datasets, Area Under Precision Recall Curve can highlight performance differences that are lost with AreaUnderRocCurve.
Remarks: The area under the ROC curve is equal to the probability that the classifier ranks a randomly chosen positive instance higher than a randomly chosen negative one (assuming 'positive' ranks higher than 'negative'). Area under the ROC curve ranges between 0 and 1, with a value closer to 1 indicating a better model.
Remarks: F1 score is the harmonic mean of precision and recall: 2 * precision * recall / (precision + recall). F1 ranges between 0 and 1, with a value of 1 indicating perfect precision and recall.
|
PercentageOfDataForTraining |
Percentage of Data for Training (0-100%) |
The percentage (0-100%) of the data to be used for training the model, the remaining percentage will be used for testing. |
BarByBarModelCreation |
Allow Bar by Bar Model Creation |
If true, when training a new model and model finds a profit target, the Deep Signal Library will only advance one bar instead of moving ahead the total of bars to target plus window size in bars. |
CreateCustomDataWindows |
Create Custom Training Data Set |
If true, allow the user to add their own data windows for training a machine learning model. This will override the existing functionality and only data windows added via DSTAddLongDataWindow, DSTAddShortDataWindow, DSTAddFailedLongDataWindow and DSTAddFailedShortDataWindow will be used in training. |
MaxAutoCreateModelTimeMins |
Create Model Time (mins) |
If true, training model will use a temporary folder to save experiment results. If false, training model will use memory to store experiment results. Note, for large data sets and longer training times this may consume large amounts of memory on your computer. |
DeleteOldDataFiles |
Delete Old Files |
If true, after a model is trained, the unnecessary data files will be deleted. This will save megabytes of hard drive space. |
DataFolder |
Strategy Data Folder |
When selecting a strategy that is derived from DSTPredictModelBase, the Deep Signal Library will display a list of machine learning models that have the same parameters added using AddDSTIndicator and AddDSTInstrument in the create model strategy as the selected predict strategy. |
MaxLongDataSets |
Max Long Data Sets |
When creating a new machine learning model, set this value to something other than 0 to set the maximum number of long profit target data sets to be used in training a model. If set to 0, the Deep Signal Library will use the total number of data sets found. |
MaxShortDataSets |
Max Short Data Sets |
When creating a new machine learning model, set this value to something other than 0 to set the maximum number of short profit target data sets to be used in training a model. If set to 0, the Deep Signal Library will use the total number of data sets found. |
MaxFailedToReachLongDataSets |
Max Failed to Reach Long Data Sets |
When creating a new machine learning model, set this value to something other than 0 to set the maximum number of failed to reach profit target data sets for long trades in training a model. If set to 0, the Deep Signal Library will use the total number of data sets found. |
MaxFailedToReachShortDataSets |
Max Failed to Reach Short Data Sets |
When creating a new machine learning model, set this value to something other than 0 to set the maximum number of failed to reach profit target data sets for short trades in training a model. If set to 0, the Deep Signal Library will use the total number of data sets found. |
UsePositiveNegativeDatasetRatio |
Use Positive Negative Data Set Ratio |
If checked, will ignore any max short or long data sets above and automatically set the max short or long data sets based on ratio. The ratio is based on Profit Target Reached/Profit Target Not Reached. |
PositiveNegativeDatasetRatio |
Positive Negative Data Set Ratio |
The ratio of Profit Target Reached/Profit Target Not Reached data sets that get set when creating a model. For example, if the ratio is set to 2, then the total Profit Target Reached data sets will be twice that of Profit Target Not Reached. |
DSTDataFolderSelection |
Data Folder Options |
The Strategy Data Folder can display all models in the Base Folder Name or models which match the same parameters as the selected strategy. |
DSTMinLongConfidenceScore |
Min Long Confidence Score (0.0-1.0) |
The minimum confidence score (0.0-1.0) for predicting a long trade. The confidence score for a trade prediction can be returned in the DSTLongTrade method. |
DSTMinShortConfidenceScore |
Min Short Confidence Score (0.0-1.0) |
The minimum confidence score (0.0-1.0) for predicting a short trade. The confidence score for a trade prediction can be returned in the DSTShortTrade method. |
Futures, foreign currency and options trading contains substantial risk and is not for every investor. An investor could potentially lose all or more than the initial investment. Risk capital is money that can be lost without jeopardizing ones financial security or lifestyle. Only risk capital should be used for trading and only those with sufficient risk capital should consider trading. Past performance is not necessarily indicative of future results.