fix kaggle

Dockerfile

@@ -1,19 +1,11 @@
-# Nasz obraz będzie dzidziczył z obrazu Ubuntu w wersji latest
 FROM ubuntu:latest
 
-# Instalujemy niezbędne zależności. Zwróć uwagę na flagę "-y" (assume yes)
 RUN apt update && apt install -y python3 \
-    python3-pip \
-    vim
+    python3-pip
 
-ENV CUTOFF=${CUTOFF}
-ENV KAGGLE_USERNAME=${KAGGLE_USERNAME}
-ENV KAGGLE_KEY=${KAGGLE_KEY}
+WORKDIR /code
 
-# Stwórzmy w kontenerze (jeśli nie istnieje) katalog /app i przejdźmy do niego (wszystkie kolejne polecenia RUN, CMD, ENTRYPOINT, COPY i ADD będą w nim wykonywane)
-WORKDIR /app
-
-# Skopiujmy nasz skrypt do katalogu /app w kontenerze
-COPY . /app/
+COPY . /code/
 
 RUN python3 -m pip install -r requirements.txt
+RUN mkdir /code/.kaggle && chmod o+w /code/.kaggle

Jenkinsfile.create_dataset

@@ -1,5 +1,5 @@
 node {
-    docker.image('s444452/ium:1.0').inside {
+    docker.image('s444452/ium:1.1').inside {
         stage('Preparation') {
             properties([
                 parameters([
@@ -31,11 +31,11 @@ node {
                 sh 'echo KAGGLE_USERNAME: $KAGGLE_USERNAME'
                 sh 'echo KAGGLE_KEY: $KAGGLE_KEY'
                 sh 'ls'
-                sh "python3 lab2_data.py"
+                sh "python3 download_dataset.py '.' 'dataset.csv'"
             }
         }
         stage('Archive artifacts') {
-            archiveArtifacts 'fake_job_postings.csv'
+            archiveArtifacts 'dataset.csv' 'train_data.csv' 'test_data.csv' 'dev_data.csv'
         }
     }
 }

Jenkinsfile.stats

@@ -1,4 +1,5 @@
 node {
+    docker.image('s444452/ium:1.1').inside {
         stage('Preparation') {
         properties([parameters([
             buildSelector(
@@ -12,15 +13,18 @@ node {
     }
     stage('Copy artifacts') {
         copyArtifacts filter: 'dataset.csv', fingerprintArtifacts: true, projectName: 's444452-create-dataset', selector: buildParameter('BUILD_SELECTOR')
+        copyArtifacts filter: 'train_data.csv', fingerprintArtifacts: true, projectName: 's444452-create-dataset', selector: buildParameter('BUILD_SELECTOR')
+        copyArtifacts filter: 'test_data.csv', fingerprintArtifacts: true, projectName: 's444452-create-dataset', selector: buildParameter('BUILD_SELECTOR')
+        copyArtifacts filter: 'dev_data.csv', fingerprintArtifacts: true, projectName: 's444452-create-dataset', selector: buildParameter('BUILD_SELECTOR')
     }
     stage('Run script') {
         withEnv(["KAGGLE_USERNAME=${params.KAGGLE_USERNAME}",
               "KAGGLE_KEY=${params.KAGGLE_KEY}"]) {
-            sh "chmod u+x ./dataset_stats.sh"
-            sh "./dataset_stats.sh"
+            sh "python3 generate_dataset_stats.py '.'"
         }
     }
     stage('Archive artifacts') {
-        archiveArtifacts 'stats.txt'
+        archiveArtifacts 'train_stats.txt' 'test_stats.txt' 'dev_stats.txt'
+    }
     }
 }

download_dataset.py

@@ -0,0 +1,60 @@
+#!/usr/bin/python
+import os.path
+import sys
+
+import pandas as pd
+from kaggle import api
+from pandas import read_csv
+from sklearn.model_selection import train_test_split
+
+
+def download_and_save_dataset(data_path, dataset_name):
+    if not os.path.exists(os.path.join(data_path, dataset_name)):
+        api.authenticate()
+        api.dataset_download_files('shivamb/real-or-fake-fake-jobposting-prediction', path=data_path,
+                                   unzip=True)
+        os.rename(os.path.join(data_path, 'fake_job_postings.csv'), os.path.join(data_path, dataset_name))
+
+
+def preprocess_dataset(data):
+    # drop columns with many nulls
+    return data.drop(['job_id', 'department', 'salary_range', 'benefits'], axis=1)
+
+
+def split_dataset(data_path, dataset_name):
+    data = read_csv(os.path.join(data_path, dataset_name))
+    data = preprocess_dataset(data)
+    train_ratio, validation_ratio, test_ratio = 0.6, 0.2, 0.2
+    data_x, data_y = data.iloc[:, :-1], data.iloc[:, -1:]
+
+    x_train, x_test, y_train, y_test = train_test_split(data_x, data_y, test_size=1 - train_ratio,
+                                                        random_state=123, stratify=data['fraudulent'])
+
+    x_val, x_test, y_val, y_test = train_test_split(x_test, y_test,
+                                                    test_size=test_ratio / (test_ratio + validation_ratio),
+                                                    random_state=123)
+
+    return x_train, x_val, x_test, y_train, y_val, y_test
+
+
+def save_dataset(data_path, data, name):
+    data.to_csv(os.path.join(data_path, name))
+
+
+def main():
+    data_path, dataset_name = sys.argv[1], sys.argv[2]
+    abs_data_path = os.path.abspath(data_path)
+    download_and_save_dataset(abs_data_path, dataset_name)
+
+    x_train, x_val, x_test, y_train, y_val, y_test = split_dataset(abs_data_path, dataset_name)
+
+    train_data = pd.concat([x_train, y_train], axis=1)
+    test_data = pd.concat([x_test, y_test], axis=1)
+    dev_data = pd.concat([x_val, y_val], axis=1)
+
+    for data, name in ((train_data, 'train_data.csv'), (test_data, 'test_data.csv'), (dev_data, 'dev_data.csv')):
+        save_dataset(abs_data_path, data, name)
+
+
+if __name__ == '__main__':
+    main()

generate_dataset_stats.py

@@ -0,0 +1,29 @@
+#!/usr/bin/python
+import os
+import pprint
+import sys
+
+from pandas import read_csv
+
+
+def save_stats_to_file(data_path, data_name, stats_name):
+    data = read_csv(os.path.join(data_path, data_name))
+    with open(os.path.join(data_path, stats_name), "w") as log_file:
+        for name, obj in (
+                ('Description: ', data.describe(include='all')), ('Shape: ', data.shape), ('Head: ', data.head())):
+            pprint.pprint(name, log_file)
+            pprint.pprint(obj, log_file)
+
+
+def main():
+    data_path = sys.argv[1]
+    abs_data_path = os.path.abspath(data_path)
+
+    for data_name, stats_name in (
+            ('train_data.csv', 'train_stats.txt'), ('test_data.csv', 'test_stats.txt'),
+            ('dev_data.csv', 'dev_stats.txt')):
+        save_stats_to_file(abs_data_path, data_name, stats_name)
+
+
+if __name__ == '__main__':
+    main()

lab2_data.py

@@ -1,36 +0,0 @@
-#!/usr/bin/python
-from kaggle import api
-from pandas import read_csv, DataFrame
-from sklearn.model_selection import train_test_split
-
-
-def download_and_save_dataset():
-    api.authenticate() 
-    api.dataset_download_files('shivamb/real-or-fake-fake-jobposting-prediction', unzip=True)
-
-
-def split_dataset(data: DataFrame):
-    train_ratio, validation_ratio, test_ratio = 0.6, 0.2, 0.2
-    data_x, data_y = data.iloc[:, :-1], data.iloc[:, -1:]
-
-    x_train, x_test, y_train, y_test = train_test_split(data_x, data_y, test_size=1 - train_ratio, random_state=123)
-
-    x_val, x_test, y_val, y_test = train_test_split(x_test, y_test,
-                                                    test_size=test_ratio / (test_ratio + validation_ratio),
-                                                    random_state=123)
-
-    return x_train, x_val, x_test, y_train, y_val, y_test
-
-
-def main():
-    # download_and_save_dataset()
-    df = read_csv('./fake_job_postings.csv')
-    print(df.describe(include='all'))
-    print(df.shape)
-    x_train, x_val, x_test, y_train, y_val, y_test = split_dataset(df)
-    print(x_train.shape, x_val.shape, x_test.shape)
-    print(y_train.shape, y_val.shape, y_test.shape)
-
-
-if __name__ == '__main__':
-    main()