# -------------------------------------------------------------------------------- NOTEBOOK-CELL: CODE
# import libraries:
# dataiku to interact with dataiku API
# pandas and numpy are default
# seaborn for plotting
# os to join paths

import dataiku
import pandas as pd, numpy as np
import seaborn as sns
import os
from dataiku import pandasutils as pdu

# -------------------------------------------------------------------------------- NOTEBOOK-CELL: CODE
# connect to the dataset using the dataiku API

iris_handle = dataiku.Dataset("iris_copy")
iris_df = iris_handle.get_dataframe()

# -------------------------------------------------------------------------------- NOTEBOOK-CELL: CODE
# connect to the folder in which we want to store the plots
# get the path to the folder with the get_path() method

# note that this will overwrite the previous version of the plot file
# if you don't want to overwrite, make the name dynamic by adding e.g. the current time to the file name

plot_folder_handle = dataiku.Folder("2qmGlaPX")
plot_folder_path = plot_folder_handle.get_path()
plot_path = os.path.join(plot_folder_path, 'iris_scatter.png')

# -------------------------------------------------------------------------------- NOTEBOOK-CELL: CODE
# create the plot and store it as a python object
# use the savefig method to save the plot

iris_pairplot = sns.pairplot(iris_df[['sepal_width', 'sepal_length', 'variety']], hue = 'variety')
iris_pairplot.savefig(plot_path)