alaiacano

apiVersion: backstage.io/v1alpha1
kind: Component
metadata:
  name: mlflow-project-example
  description: A collection of MLFlow experiments.
  annotations:
    mlflow/experiment: '1'
  tags:
    - machine-learning
spec:

alaiacano

import com.twitter.algebird._
import com.twitter.algebird.Operators._

import scala.util.Random
Random.setSeed(0L)

/**
  * An "arm" represents an option in our A/B test.
  *
  * @param name Name of this option

alaiacano

Blink 182 and Green Day were and are still very popular bands. Therefore, they'll end up on a lot of playlists. Here are some genres where it would make sense to see them. I'll assign a distribution too, since obviously a band isn't limited to one kind of playlist.

So let's say Green Day is on 2,000 different playlists, broken down among these genres:

• Punk - 35%
• Rock - 20%
• 90s - 20%
• Alternative - 20%
• Party - 5%

alaiacano

record FirstType {
    union { null, int } field1 = null;
    union { null, int } field2 = null;
}

record SecondType {
    union { null, long } someOtherField = null;
}

record MyAvroObjec {

alaiacano

  // load the data and convert it to a Glass object.
  val data: TypedPipe[Glass] = TextLine(args("input"))
    .map{line => Glass.fromTextLine(line)}

  // open the SVM model used to predict the type of glass.
  val svmModel = getClass.getResource(modelFile).openStream()
  val svmModelEvaluator = Predictable.buildEvaluator(svmModel)

  val predictedType: TypedPipe[(Int, GlassType.PredictType)] = data
    // turn the Glass object into a GlassType

alaiacano

resolvers ++= Seq(
    "Conjars" at "http://conjars.org/repo"
)

libraryDependencies ++= Seq(
  "org.jpmml" % "pmml-evaluator" % "1.0.21",
  "org.jpmml" % "pmml-model" % "1.0.21"
)

alaiacano

import com.twitter.scalding._
import TDsl._

class MapTest(args: Args) extends Job(args) {
  val data = TypedTsv[(Int, Int, String, String)](args("input"))
    .map{ case (a, b, c, d) => 
      val e = if (a > b) c else d
      (a, b, c, d, e)
    }
    .write(TypedTsv[(Int, Int, String, String, String)](args("output")))

alaiacano

import com.twitter.scalding._
import scala.collection.mutable.ListBuffer
import TDsl._

/*
INPUT:
a	1
a	2
a	3
b	1

alaiacano

class NBTestJob(args: Args) extends Job(args) {

  val output = args("output")

  val iris = Tsv("iris.tsv", ('id, 'class, 'sepalLength, 'sepalWidth, 'petalLength, 'petalWidth))
    .read

  // The model expects data in "melted" form.
  // One Field for the variable names, one for the coresponding values.
  val irisMelted = iris

alaiacano

import numpy as np
from pylab import *
import scipy.signal as signal

fs = 1 # 1 day sampling freq

filt = np.ones(7) / 7

w, resp = signal.freqz(filt)