Note:

Tested on Python 2.7 with spark 2.0

• Run the below line of code to install 0.18 of scikitlearn

!pip install scikit-learn==0.18

Classification

• Classification classification-12Liner consists of a shortened version of IRIS classifier with no evaluation
• Classification Simple outputs a related image basing on the classification of the flower
• Classification_Evaluation contains the Complete version of Basic IRIS data classification along with performance evaluation and algorithm comparision
• MNIST Simple classifies random test images of hand written digits from MNIST Dataset into respective classes of 0-9

Regression

• Regression_realestate uses Linear regression to predict Real estate prices based on a dataset included in RealestateData.csv
• Regression_tv predicts which of the two Tv series flash or Arrow gets most veiwership based on the data from veiwershipData.csv

Importing Notebooks onto DSX

• Navigate to https://datascience.ibm.com/ and Login with your IBM ID
• Create your project and Give it a Description
• Click on Add notebooks

From From URL

• Copy the URL of the python notebook File
• Click on From URL Tab
• Give your project a name and paste the copied url in Notebook URL Field

From File

• Download the gitHub Repo
• Click on From File Tab
• Give your notebook a name and click browse button to import your .ipnyb file

Accessing DataFiles

• Click on add data assets from the project home screen
• Import your .csv data file
• From within the notebook, click the Find and add Data menu item on the top right
• Find your dataset and select Insert Pandas Dataframe
• use the assigned Data variable which can be seen in the second-bottom most line of the inserted code