This is a project in Large scale data processing to classify the damaged and non-damaged cars using distributed SVM in Pyspark and Hadoop
-
Updated
Jun 10, 2021 - Jupyter Notebook
This is a project in Large scale data processing to classify the damaged and non-damaged cars using distributed SVM in Pyspark and Hadoop
To associate your repository with the distributed-svm topic, visit your repo's landing page and select "manage topics."