import numpy as np
from numpy.random import randn
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
from mpld3 import display_d3

np.random.seed(0)

def sinplot(flip=1):
    x = np.linspace(0, 14, 100)
    for i in range(1, 7):
        plt.plot(x, np.sin(x + i * .5) * (7 - i) * flip)

x1 = randn(80)
x2 = randn(80)
x3 = x1 * x2
y1 = .5 + 2 * x1 - x2 + 2.5 * x3 + 3 * randn(80)
y2 = .5 + 2 * x1 - x2 + 2.5 * randn(80)
y3 = y2 + randn(80)
g = np.random.choice(list("ABC"), 80)
y_logistic = 1 / (1 + np.exp(-y1))
y_flip = [np.random.binomial(1, p) for p in y_logistic]
df = pd.DataFrame(dict(x1=x1, x2=x2, x3=x3, y1=y1, y2=y2, y3=y3, y_flip=y_flip, g=g))
d = randn(100, 30)

sinplot()

sinplot()
display_d3()

sns.set(style="whitegrid")
sinplot()

sinplot()
display_d3()

sns.set(style="nogrid")
sinplot()

sinplot()
display_d3()

sns.set()

sns.lmplot("x1", "y2", df)
fig = plt.gcf()

display_d3(fig)

sns.lmplot("x1", "y2", df, col="g", size=3.5)
fig = plt.gcf()

display_d3(fig)

sns.lmplot("x1", "y2", df, color="g")
fig = plt.gcf()

display_d3(fig)

sns.regplot("x1", "y1", df)
fig = plt.gcf()

display_d3(fig)

sns.coefplot("y1 ~ x1 * x2 + g", df)
fig = plt.gcf()

display_d3(fig)

sns.interactplot("x1", "x2", "y1", df)
fig = plt.gcf()

display_d3(fig)

sns.corrplot(df)
fig = plt.gcf()

display_d3(fig)

sns.boxplot(df.x1, df.g)
fig = plt.gcf()

display_d3(fig)

sns.violinplot(df.x1, df.g)
fig = plt.gcf()

display_d3(fig)

sns.distplot(df.x1)
fig = plt.gcf()

display_d3(fig)

sns.kdeplot(df.x1, shade=True)
fig = plt.gcf()

display_d3(fig)

sns.kdeplot(df[["x1", "x2"]])
fig = plt.gcf()

display_d3(fig)

sns.kdeplot(df[["x1", "x2"]], shade=True)
fig = plt.gcf()

display_d3(fig)

def random_walk(n, start=0, p_inc=.2):
    return start + np.cumsum(np.random.uniform(size=n) < p_inc)
starts = np.random.choice(range(4), 10)
probs = [.1, .3, .5]
walks = np.dstack([[random_walk(15, s, p) for s in starts] for p in probs])

sns.tsplot(walks)
fig = plt.gcf()

display_d3(fig)

sns.tsplot(walks, err_style="ci_bars")
fig = plt.gcf()

display_d3(fig)

def sine_wave(n_x, obs_err_sd=1.5, tp_err_sd=.3):
    x = np.linspace(0, (n_x - 1) / 2, n_x)
    y = np.sin(x) + np.random.normal(0, obs_err_sd) + np.random.normal(0, tp_err_sd, n_x)
    return y
sines = np.array([sine_wave(31) for _ in range(20)])

sns.tsplot(sines, err_style="boot_traces", n_boot=500)
fig = plt.gcf()

display_d3(fig)

sns.tsplot(sines, err_style="boot_kde", n_boot=500)
fig = plt.gcf()

display_d3(fig)