| Year/Month | BEV | sales | |
|---|---|---|---|
| 2011/01 | 588 | 310 | |
| 2011/02 | 47 | 417 | |
| 2011/03 | 159 | 286 | |
| 2011/04 | 173 | 162 | |
| 2011/05 | 305 | 470 | |
| 2011/06 | 319 | 273 | |
| 2011/07 | 535 | 650 | |
| 2011/08 | 917 | 1323 | |
| 2011/09 | 57 | 688 |
| from numpy.random import seed | |
| from numpy.random import randn | |
| from scipy.stats import shapiro | |
| ''' | |
| 24.5.2 Shapiro-Wilk Test | |
| The Shapiro-Wilk test evaluates a data sample and quanti es how likely it is that the data | |
| was drawn from a Gaussian distribution | |
| ''' | |
| # normality test | |
| stat, p = shapiro(sample_singles) |
| from statsmodels.graphics.gofplots import qqplot | |
| from matplotlib import pyplot | |
| # q-q plot | |
| sample_normal = normal(size=1000) | |
| qqplot(sample_normal, line='s') | |
| qqplot(sample, line='s') | |
| pyplot.show() |
| import numpy as np | |
| not_nan_mask = np.isnan(np.array(bank_product_rec_santander_data.renta.values))==False | |
| renta_values_not_nan=bank_product_rec_santander_data.renta.values[not_nan_mask] | |
| sub_renta_mean = np.mean(renta_values_not_nan).round(2) | |
| bank_product_rec_santander_sub_df['renta'] = bank_product_rec_santander_sub_df['renta'].fillna(value=sub_renta_mean) |
| import plotly.graph_objects as go | |
| fig = go.Figure(data=go.Scatter(x=dataset.index, y=dataset.dependent_variable, mode='lines+markers+text')) | |
| fig.show() |
| def makeCartesianProduct(array_x, array_y, array_z): | |
| import pandas as pd | |
| import itertools | |
| return pd.DataFrame.from_records(itertools.product(array_x.reshape(-1, ), array_y.reshape(-1, ), array_z.reshape(-1, )), | |
| columns=['x', 'y', 'z']) | |
| combinations_cartesian_prod = makeCartesianProduct(jobs_df.A.unique(), jobs_df.B.unique(), jobs_df.C.unique()) |
| # split a univariate dataset into train/test sets | |
| def train_test_split(data, n_test): | |
| return data[:-int(n_test)], data[-int(n_test):] | |
| # transform list into supervised learning format | |
| def series_to_supervised(data, n_in, n_out=1): | |
| import pandas as pd | |
| df = pd.DataFrame(data) | |
| cols = list() |
| # bootstrap confidence intervals | |
| import numpy as np | |
| from numpy.random import seed, rand, randint, randn, std | |
| from numpy import mean, median, percentile | |
| from scipy.stats import linregress | |
| # seed random number generator | |
| seed(1) | |
| # prepare data | |
| x = 20 * randn(1000) + 100 |
People
 :bowtie: |
😄 :smile: |
😆 :laughing: |
|---|---|---|
😊 :blush: |
😃 :smiley: |
:relaxed: |
😏 :smirk: |
😍 :heart_eyes: |
😘 :kissing_heart: |
😚 :kissing_closed_eyes: |
😳 :flushed: |
😌 :relieved: |
😆 :satisfied: |
😁 :grin: |
😉 :wink: |
😜 :stuck_out_tongue_winking_eye: |
😝 :stuck_out_tongue_closed_eyes: |
😀 :grinning: |
😗 :kissing: |
😙 :kissing_smiling_eyes: |
😛 :stuck_out_tongue: |
| #source: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.Series.interpolate.html | |
| def imputeMissingInterpolating(array_df): | |
| import numpy as np | |
| for attr in array_df.columns: | |
| attribute_interpolated = array_df[attr].interpolate(method='linear', limit_direction='both') | |
| assert len(attribute_interpolated[np.isnan(attribute_interpolated)]) == 0 | |
| array_df[attr] = attribute_interpolated | |
| return array_df |