DOC update gallery examples to load datasets from path (skrub-data#1940)

Author: MuditAtrey

CHANGES.rst

@@ -53,6 +53,12 @@ Bug Fixes
 - Errors raised when a polars LazyFrame is passed where an eager DataFrame is
   expected are now clearer. :pr:`1916` by :user:`Jérôme Dockès <jeromedockes>`.
 
+Documentation
+-------------
+- Updated gallery examples to load datasets from their file paths using
+  ``pd.read_csv()``, following the pattern established in :pr:`1852`.
+  :pr:`1940` by :user:`MuditAtrey <MuditAtrey>`.
+
 Release 0.7.2
 =============
 

examples/0040_fuzzy_joining.py

@@ -34,10 +34,12 @@
 # --------------------------------
 #
 # We import the happiness score table first:
+import pandas as pd
+
 from skrub import datasets
 
 happiness_data = datasets.fetch_country_happiness()
-df = happiness_data.happiness_report
+df = pd.read_csv(happiness_data.happiness_report_path)
 
 ###############################################################################
 # Let's look at the table:
@@ -65,17 +67,17 @@
 # <https://data.worldbank.org/>`_, which are also available in the dataset
 # We extract the table containing GDP per capita by country:
 
-gdp_per_capita = happiness_data.GDP_per_capita
+gdp_per_capita = pd.read_csv(happiness_data.GDP_per_capita_path)
 gdp_per_capita.head(3)
 
 ###############################################################################
 # Then another table, with life expectancy by country:
-life_exp = happiness_data.life_expectancy
+life_exp = pd.read_csv(happiness_data.life_expectancy_path)
 life_exp.head(3)
 
 ###############################################################################
 # And a table with legal rights strength by country:
-legal_rights = happiness_data.legal_rights_index
+legal_rights = pd.read_csv(happiness_data.legal_rights_index_path)
 legal_rights.head(3)
 
 ###############################################################################

examples/0060_multiple_key_join.py

@@ -49,7 +49,7 @@
 
 dataset = fetch_flight_delays()
 seed = 1
-flights = dataset.flights
+flights = pd.read_csv(dataset.flights_path)
 
 # Sampling for faster computation.
 flights = flights.sample(5_000, random_state=seed, ignore_index=True)
@@ -76,7 +76,7 @@
 #     - The ``airports`` dataset, with information such as their name
 #       and location (longitude, latitude).
 
-airports = dataset.airports
+airports = pd.read_csv(dataset.airports_path)
 airports.head()
 
 ########################################################################
@@ -86,7 +86,7 @@
 #       Both tables are from the Global Historical Climatology Network.
 #       Here, we consider only weather measurements from 2008.
 
-weather = dataset.weather
+weather = pd.read_csv(dataset.weather_path)
 # Sampling for faster computation.
 weather = weather.sample(10_000, random_state=seed, ignore_index=True)
 weather.head()
@@ -95,7 +95,7 @@
 #     - The ``stations`` dataset. Provides location of all the weather
 #       measurement stations in the US.
 
-stations = dataset.stations
+stations = pd.read_csv(dataset.stations_path)
 stations.head()
 
 ###############################################################################

examples/0070_join_aggregation.py

@@ -80,11 +80,15 @@
 """
 
 # %%
+import pandas as pd
+
 from skrub import TableReport
 from skrub.datasets import fetch_credit_fraud
 
 bunch = fetch_credit_fraud()
-products, baskets = bunch.products, bunch.baskets
+products = pd.read_csv(bunch.products_path)
+baskets = pd.read_csv(bunch.baskets_path)
+
 TableReport(products)
 
 # %%
@@ -111,7 +115,6 @@
 # Then, we can expand all lists into columns, as if we were "flattening" the dataframe.
 # We end up with a products dataframe ready to be joined on the baskets dataframe, using
 # ``"basket_ID"`` as the join key.
-import pandas as pd
 
 products_flatten = []
 for col in products_grouped.columns:

examples/0080_interpolation_join.py

@@ -40,9 +40,9 @@
 from skrub.datasets import fetch_flight_delays
 
 dataset = fetch_flight_delays()
-weather = dataset.weather
+weather = pd.read_csv(dataset.weather_path)
 weather = weather.sample(100_000, random_state=0, ignore_index=True)
-stations = dataset.stations
+stations = pd.read_csv(dataset.stations_path)
 weather = stations.merge(weather, on="ID")[
     ["LATITUDE", "LONGITUDE", "YEAR/MONTH/DAY", "TMAX", "PRCP", "SNOW"]
 ]
@@ -128,11 +128,13 @@
 # ``'Origin'`` which refers to the departure airport’s IATA code. We use only a subset
 # to speed up the example.
 
-flights = dataset.flights
+flights = pd.read_csv(dataset.flights_path)
 flights["Year_Month_DayofMonth"] = pd.to_datetime(flights["Year_Month_DayofMonth"])
 flights = flights[["Year_Month_DayofMonth", "Origin", "ArrDelay"]]
 flights = flights.sample(20_000, random_state=0, ignore_index=True)
-airports = dataset.airports[["iata", "airport", "state", "lat", "long"]]
+airports = pd.read_csv(dataset.airports_path)[
+    ["iata", "airport", "state", "lat", "long"]
+]
 flights = flights.merge(airports, left_on="Origin", right_on="iata")
 # printing the first row is more readable than the head() when we have many columns
 flights.iloc[0]