story_bike_share_analyze_11.py

# Import the weather data and drop the first 22 rows (containing descriptions of the weather station)
weather = pd.read_csv('./data/weather.csv', header=22)

# Remove units contained in the column names (eg. Celcius, mm, etc.)
weather.columns = [re.sub(r'\([^()]*\)', '', x).strip() if x != 'Date/Time' else 'Date' for x in weather.columns]

data = df.groupby(['Date', 'User Type'])['Id'].nunique().to_frame().pivot_table(index='Date', columns='User Type').reset_index()
data.columns = ['Date', 'Casual Trips', 'Member Trips']
data = data.merge(weather[['Date', 'Mean Temp', 'Total Precip']], on='Date', how='inner')

g = sns.pairplot(data, diag_kind='kde', plot_kws={'s': 10})
g.fig.set_size_inches(10, 10)

fig, ax = plt.subplots(figsize=(12, 10))

corr = data.corr()
mask = np.zeros_like(corr)
mask[np.triu_indices_from(mask)] = True

sns.heatmap(corr, mask=mask, annot=True, cmap=sns.diverging_palette(240, 10, as_cmap=True), center=0)