Refactor_SpeedOtimize.txt

1. Using "for...in..."
-----------------------------------------------
From:
new_feature_data = feature_data.rename(columns={'fixed acidity': 'fixed_acidity',
                                                'volatile acidity': 'volatile_acidity',
                                                'citric acid': 'citric_acid',
                                                'residual sugar': 'residual_sugar',
                                                'free sulfur dioxide': 'free_sulfur_dioxide',
                                                'total sulfur dioxide': 'total_sulfur_dioxide'
                                                })

To:
labels = list(feature_data.columns)
labels[:] = [label.replace(' ', '_') for label in labels]
feature_data.columns = labels

feature_data.head()
-----------------------------------------------

2. Using def funct (avoid repeating steps of 'alcohol' to 'pH', 'residual_sugar'...)
-----------------------------------------------
From:
median_alcohol = feature_data.alcohol.median()
for i, alcohol in enumerate(feature_data.alcohol):
    if alcohol >= median_alcohol:
        feature_data.loc[i, 'alcohol'] = 'high'
    else:
        feature_data.loc[i, 'alcohol'] = 'low'
feature_data.groupby('alcohol').quality.mean()

To:
def median_val(feature_data, feature):
    median = getattr(getattr(feature_data, feature), "median")
    median = median()
    for i, feature_name in enumerate(getattr(feature_data, feature)):
        if feature_name >= median:
            feature_data.loc[i, feature] = 'high'
        else:
            feature_data.loc[i, feature] = 'low'
    print(feature_data.groupby(feature).quality.mean(), '\n')

median_val(feature_data, 'alcohol')

Or: ==> only works for "pandas array" NOT numpy or python array
def median_val(feature_data, feature):
    median = feature_data[feature].median()
    for i, feature_name in enumerate(feature_data[feature]):
        if feature_name >= median:
            feature_data.loc[i, feature] = 'high'
        else:
            feature_data.loc[i, feature] = 'low'
    print(feature_data.groupby(feature).quality.mean(), '\n')

median_val(feature_data, 'alcohol')
-----------------------------------------------

3. Optimizing speed:
-----------------------------------------------
From:
for book in recent_books:
    if book in coding_books:
        recent_coding_books.append(book)

To:
numpy.intersect1d(recent_books, coding_books)

Or:
set(coding_books).intersection(recent_books)
-----------------------------------------------
From:
for cost in gift_costs:
    if cost < 25:
        total_price += cost * 1.08  # add cost after tax

To: 
total_price =  np.sum(gift_costs[gift_costs < 25]) * 1.08
-----------------------------------------------
From:
for x in list:
    sum += x*x - y

To: 
sum((x**2 - y) for x in list)