Added CUTOFF parameter

Jenkinsfile

@@ -12,6 +12,11 @@ pipeline {
             description: 'Kaggle token taken from kaggle.json file, as described in https://github.com/Kaggle/kaggle-api#api-credentials',
             name: 'KAGGLE_KEY'
         )
+        string(
+            defaultValue: "260000",
+            description: 'Use CUTOFF random examples from dataset',
+            name: 'CUTOFF'
+        )
     }
     stages {
         stage('Hello') {
@@ -27,8 +32,9 @@ pipeline {
         stage("Download data") {
             steps {
                 withEnv(["KAGGLE_USERNAME=${params.KAGGLE_USERNAME}",
-                         "KAGGLE_KEY=${params.KAGGLE_KEY}"]) {
-                    sh "./download_data.sh"
+                         "KAGGLE_KEY=${params.KAGGLE_KEY}",
+                         "CUTOFF=%{params.CUTOFF}"]) {
+                    sh "./download_data.sh $CUTOFF"
                     archiveArtifacts artifacts: "crime.test, crime.dev, crime.train", onlyIfSuccessful: true
                 }
             }

download_data.sh

@@ -2,7 +2,7 @@
 
 kaggle datasets download -d AnalyzeBoston/crimes-in-boston
 unzip crimes-in-boston.zip
-shuf crime.csv > crime.shuf
+shuf crime.csv | head -n $1 > crime.shuf
 head -n 30000 crime.shuf > crime.test
 head -n 60000 crime.shuf | tail -n 30000 > crime.dev
 tail -n +60001 crime.shuf > crime.train