Docs can be found at: https://hntd187.github.io/scalagen/docs/0.0.2/
Scala gen is a simple library based around being able to generate row based test data and easily compose a description of that data set. It also currently includes the ability to write this data to both Csv and Parquet files.
"com.scalagen" %% "scalagen" % "0.0.2" % Test
You then can compose a simple data set like so
import com.scalagen.data._
val data = IncrementingSource() | YesNoSource()
data.getLine- GaussianSource (Numbers drawn from a normal distribution)
- BernoulliSource (Booleans drawn from a Bernoulli Trial)
- GenderSource (M or F based on a Bernoulli Trial)
- YesNoSource (Y or N based on a Bernoulli Trial)
- DateSource (Incrementing Dates)
- IncrementingSource (Incrementing integers, like a primary key)
- RandomSource (Random values drawn from a provided set of values)
- Simple CSV
- Simple Parquet
import com.scalagen.data._
val data = Csv {
IncrementingSource() | YesNoSource() | GaussianSource()
}
data.write("data.csv", 10)
data.show()Produces the following
╒════════════════════════════════╤════════════════════════════════╤════════════════════════════════╕
│ Id │ Response │ Score │
╞════════════════════════════════╪════════════════════════════════╪════════════════════════════════╡
│ 1 │ No │ 4.7 │
│ 2 │ Yes │ 0.85 │
│ 3 │ No │ -0.55 │
│ 4 │ Yes │ 0.37 │
│ 5 │ Yes │ 1.0 │
│ 6 │ No │ 2.3 │
│ 7 │ No │ 4.0 │
│ 8 │ Yes │ 1.4 │
│ 9 │ Yes │ 1.1 │
│ 10 │ Yes │ 0.92 │
╘════════════════════════════════╧════════════════════════════════╧════════════════════════════════╛
You can also alter the format of Doubles and Dates by providing implicit values for those...
import com.scalagen.data._
implicit val mc = MathContext.DECIMAL32
implicit val df = DateTimeFormatter.BASIC_ISO_DATE
val data = Csv {
IncrementingSource() | DateSource(LocalDate.now()) | GaussianSource()
}.withHeaders("Id", "Date", "Score")
data.show()Which produces...
╒════════════════════════════════╤════════════════════════════════╤════════════════════════════════╕
│ Id │ Date │ Score │
╞════════════════════════════════╪════════════════════════════════╪════════════════════════════════╡
│ 1 │ 20161012 │ 2.102348 │
│ 2 │ 20161019 │ 1.360111 │
│ 3 │ 20161026 │ 2.734621 │
│ 4 │ 20161102 │ 1.824946 │
│ 5 │ 20161109 │ 0.9169786 │
│ 6 │ 20161116 │ 2.418038 │
│ 7 │ 20161123 │ 2.647185 │
│ 8 │ 20161130 │ 2.361134 │
│ 9 │ 20161207 │ 2.786133 │
│ 10 │ 20161214 │ 3.19905 │
╘════════════════════════════════╧════════════════════════════════╧════════════════════════════════╛
=======
Scala gen is a simple library based around being able to generate row based test data and easily compose a description of that data set. It also currently includes the ability to write this data to both Csv and Parquet files.
"com.scalagen" %% "scalagen" % "0.0.2" % Test
You then can compose a simple data set like so
import com.scalagen.data._
val data = IncrementingSource() | YesNoSource()
data.getLine- GaussianSource (Numbers drawn from a normal distribution)
- BernoulliSource (Booleans drawn from a Bernoulli Trial)
- GenderSource (M or F based on a Bernoulli Trial)
- YesNoSource (Y or N based on a Bernoulli Trial)
- DateSource (Incrementing Dates)
- IncrementingSource (Incrementing integers, like a primary key)
- RandomSource (Random values drawn from a provided set of values)
- Simple CSV
- Simple Parquet
import com.scalagen.data._
val data = Csv {
IncrementingSource() | YesNoSource() | GaussianSource()
}
data.write("data.csv", 10)
data.show()Produces the following
╒════════════════════════════════╤════════════════════════════════╤════════════════════════════════╕
│ Id │ Response │ Score │
╞════════════════════════════════╪════════════════════════════════╪════════════════════════════════╡
│ 1 │ No │ 4.7 │
│ 2 │ Yes │ 0.85 │
│ 3 │ No │ -0.55 │
│ 4 │ Yes │ 0.37 │
│ 5 │ Yes │ 1.0 │
│ 6 │ No │ 2.3 │
│ 7 │ No │ 4.0 │
│ 8 │ Yes │ 1.4 │
│ 9 │ Yes │ 1.1 │
│ 10 │ Yes │ 0.92 │
╘════════════════════════════════╧════════════════════════════════╧════════════════════════════════╛
You can also alter the format of Doubles and Dates by providing implicit values for those...
import com.scalagen.data._
implicit val mc = MathContext.DECIMAL32
implicit val df = DateTimeFormatter.BASIC_ISO_DATE
val data = Csv {
IncrementingSource() | DateSource(LocalDate.now()) | GaussianSource()
}.withHeaders("Id", "Date", "Score")
data.show()Which produces...
╒════════════════════════════════╤════════════════════════════════╤════════════════════════════════╕
│ Id │ Date │ Score │
╞════════════════════════════════╪════════════════════════════════╪════════════════════════════════╡
│ 1 │ 20161012 │ 2.102348 │
│ 2 │ 20161019 │ 1.360111 │
│ 3 │ 20161026 │ 2.734621 │
│ 4 │ 20161102 │ 1.824946 │
│ 5 │ 20161109 │ 0.9169786 │
│ 6 │ 20161116 │ 2.418038 │
│ 7 │ 20161123 │ 2.647185 │
│ 8 │ 20161130 │ 2.361134 │
│ 9 │ 20161207 │ 2.786133 │
│ 10 │ 20161214 │ 3.19905 │
╘════════════════════════════════╧════════════════════════════════╧════════════════════════════════╛
