From 1ac6cebc4138185156a60c3368e83fc3b30d4ebe Mon Sep 17 00:00:00 2001 From: s444439 Date: Wed, 19 Apr 2023 20:22:37 +0200 Subject: [PATCH] update --- Jenkinsfile | 3 +-- create-dataset.py | 4 ++-- 2 files changed, 3 insertions(+), 4 deletions(-) diff --git a/Jenkinsfile b/Jenkinsfile index 16dd15d..8d176c7 100644 --- a/Jenkinsfile +++ b/Jenkinsfile @@ -19,8 +19,7 @@ pipeline { steps { sh 'git clone https://git.wmi.amu.edu.pl/s444439/ium_z444439' sh 'curl -O https://archive.ics.uci.edu/ml/machine-learning-databases/adult/adult.data' - sh 'sed -i "1i\\age,workclass,fnlwgt,education,education-num,marital-status,occupation,relationship,race,sex,capital-gain,capital-loss,hours-per-week,native-country,income" adult.data > adult.csv' - sh 'rm adult.data' + sh 'sed "1i\\age,workclass,fnlwgt,education,education-num,marital-status,occupation,relationship,race,sex,capital-gain,capital-loss,hours-per-week,native-country,income" adult.data > adult.csv'sh 'rm adult.data' sh 'ls -a' sh 'ls -a ./ium_z444439' diff --git a/create-dataset.py b/create-dataset.py index 6b630fb..a233a6d 100644 --- a/create-dataset.py +++ b/create-dataset.py @@ -1,9 +1,9 @@ -import pandas +import pandas as pd import os from sklearn.model_selection import train_test_split CUTOFF = int(os.environ['CUTOFF']) -adults = pandas.read_csv('adult.csv') +adults = pd.read_csv('adult.csv') adults = adults.dropna() adults = adults.sample(CUTOFF)