From ea3b98f2ef3673a89e6e9dc4d9572d8604cc2ea2 Mon Sep 17 00:00:00 2001 From: wojciechbatruszewicz Date: Mon, 26 Jun 2023 19:04:04 +0200 Subject: [PATCH] Create dataset --- Jenkinsfile | 2 +- createDataset.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/Jenkinsfile b/Jenkinsfile index 8dfdd38..268d4e3 100644 --- a/Jenkinsfile +++ b/Jenkinsfile @@ -28,7 +28,7 @@ pipeline { withEnv(["KAGGLE_USERNAME=${params.KAGGLE_USERNAME}", "KAGGLE_KEY=${params.KAGGLE_KEY}" ]) { sh 'kaggle datasets download -d elakiricoder/gender-classification-dataset' - sh 'unzip -o gender_classification_v7.zip' + sh 'unzip -o gender-classification-dataset.zip' } } } diff --git a/createDataset.py b/createDataset.py index 92a3558..2cbe09e 100644 --- a/createDataset.py +++ b/createDataset.py @@ -1,7 +1,7 @@ import pandas as pd from sklearn.preprocessing import MinMaxScaler from sklearn.model_selection import train_test_split -gender_classification = pd.read_csv('gender_classification_v7.csv.csv') +gender_classification = pd.read_csv('gender_classification_v7.csv') gender_classification_train_final, gender_classification_test = train_test_split(gender_classification, test_size=0.2, random_state=1) gender_classification_test_final, gender_classification_val_final = train_test_split(gender_classification_test, test_size=0.5, random_state=1)