Dataset Generator

---

Info:

This a dataset generator that generates data for use in data science.

The data is for use mainly in regression analysis. The data in both columns is a Fahrenheit column and a Celsius column.

The arrangement of the columns is dependent on the fields selected before pressing Generate Data button.

The two main Feature of this simple program i to generate a sample training data and testing data. The validation data can be split from training data.

---

Running the executable

clone the repository by running the code

git clone https://github.com/Niccher/Data-Generator.git

cd Data-Generator/dist

Note for the code to be executed , you need to first have Java installed

java -jar Data_Generator.jar

---

Generating Training data

A resulting window will appear where you will interact with the client

The selected tab Training Data generates data for use as the training data, has several features I have assigned to it such as ,

1. Category: This has the format of the output, that is the order of columns in the output datafile.

2. Output Format: The extension of the file and the delimiter between the columns per row. These delimiters can only be either a comma or a tab.

3. Batch Size:

   This specifies the number of rows one wants to generate in the output file.

4. Output Location Indicator:

   The Browse button opens a file-chooser prompt that asks you where you want to save the data.

To generate a dataset ensure you select all the above feature then initiate the generator by pressing the Start Generating button.

A confirmation window shall pop-up to indicate that it's finished, Simply as that your training data has been generated and saved for you.

---

Generating Testing data

To generate testing data, press the next tab and a new panel will appear

The selected tab Testing Data generates data for use as the testing data, it has ,

1. Category: This has the format of the output, that is the order of columns in the output datafile.

2. Batch Size:

   This specifies the number of rows one wants to generate in the output file.

3. Output Location Indicator:

   The Browse button opens a file-chooser prompt that asks you where you want to save the data.

   ---

   Algorithms Used

4. Random Number Generator

   The application uses a simple random generator that . The method implementing this is bounded by the length of number generated

   private int getRandomNumberInRange(int min, int max) {
   	Random r = new Random();
       return r.ints(min, (max + 1)).limit(1).findFirst().getAsInt();
    }

   This method is invoked till the all iterations are completed. The iterations are specified by the batch number you specified earlier, as shown below.

   

5. Converters

   The temperature converter [Fahrenheit to Celsius and vice versa] is enabled by using this two methods.

   • Fahrenheit to Celsius

     private float FahtoCelsius(float parsed){
         //T(°C) = (T(°F) - 32) / (1.8)
         float cel=(float) ((parsed - 32) / (1.8));
         return cel;
        }

   • Celsius to Fahrenheit

     private float CelsiustoFah(float parsed){
         //F = C * 1.8 + 32  
         float fah=(float) ((parsed * 1.8) + 32);
         return fah;
     }