Commit ae2830e

committed

提交 seaborn-3 图片和代码示例

1 parent 5326cd8 commit ae2830eCopy full SHA for ae2830e

File tree

21 files changed

+103

-0

lines changed

day-126

21 files changed

+103

-0

lines changed

`‎day-126/2.py`

Lines changed: 54 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,54 @@`
	`1`	`+import seaborn as sns`
	`2`	`+import matplotlib.pyplot as plt`
	`3`	`+from scipy import stats`
	`4`	`+sns.set(color_codes=True)`
	`5`	`+`
	`6`	`+# 2.1 示例`
	`7`	`+x = np.random.normal(size=100)`
	`8`	`+sns.distplot(x)`
	`9`	`+`
	`10`	`+# 2.2 示例`
	`11`	`+x = np.random.normal(size=100)`
	`12`	`+sns.distplot(x, kde=False, rug=True)`
	`13`	`+`
	`14`	`+# 2.3 示例`
	`15`	`+x = np.random.normal(size=100)`
	`16`	`+sns.distplot(x, hist=False, rug=True)`
	`17`	`+`
	`18`	`+x = np.random.normal(0, 1, size=30) # 初始化一组服从正态分布的随机数`
	`19`	`+bandwidth = 1.06 * x.std() * x.size ** (-1 / 5.) # 根据经验公式计算 KDE 的带宽`
	`20`	`+support = np.linspace(-4, 4, 200)`
	`21`	`+`
	`22`	`+kernels = []`
	`23`	`+for x_i in x:`
	`24`	`+`
	`25`	`+ kernel = stats.norm(x_i, bandwidth).pdf(support) # 获取每一个观测值的核密度估计`
	`26`	`+ kernels.append(kernel)`
	`27`	`+ plt.plot(support, kernel, color="r") # 为每一个观测值绘制核密度估计曲线`
	`28`	`+`
	`29`	`+sns.rugplot(x, color=".2", linewidth=3)`
	`30`	`+`
	`31`	`+from scipy.integrate import trapz`
	`32`	`+density = np.sum(kernels, axis=0)`
	`33`	`+density /= trapz(density, support)`
	`34`	`+plt.plot(support, density)`
	`35`	`+`
	`36`	`+sns.kdeplot(x, shade=True)`
	`37`	`+`
	`38`	`+sns.kdeplot(x)`
	`39`	`+sns.kdeplot(x, bw=.2, label="bw: 0.2")`
	`40`	`+sns.kdeplot(x, bw=2, label="bw: 2")`
	`41`	`+plt.legend()`
	`42`	`+`
	`43`	`+sns.kdeplot(x, shade=True, cut=0)`
	`44`	`+sns.rugplot(x)`
	`45`	`+`
	`46`	`+# 2.4 示例`
	`47`	`+x = np.random.gamma(6, size=200)`
	`48`	`+sns.distplot(x, kde=False, fit=stats.gamma)`
	`49`	`+`
	`50`	`+`
	`51`	`+`
	`52`	`+`
	`53`	`+`
	`54`	`+`

`‎day-126/3.py`

Lines changed: 38 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,38 @@`
	`1`	`+import seaborn as sns`
	`2`	`+import matplotlib.pyplot as plt`
	`3`	`+from scipy import stats`
	`4`	`+sns.set(color_codes=True)`
	`5`	`+`
	`6`	`+mean, cov = [0, 1], [(1, .5), (.5, 1)]`
	`7`	`+data = np.random.multivariate_normal(mean, cov, 200)`
	`8`	`+df = pd.DataFrame(data, columns=["x", "y"])`
	`9`	`+`
	`10`	`+# 3.1 示例`
	`11`	`+sns.jointplot(x="x", y="y", data=df)`
	`12`	`+`
	`13`	`+# 3.2 示例`
	`14`	`+x, y = np.random.multivariate_normal(mean, cov, 1000).T`
	`15`	`+with sns.axes_style("white"):`
	`16`	`+ sns.jointplot(x=x, y=y, kind="hex", color="k")`
	`17`	`+`
	`18`	`+# 3.3 示例`
	`19`	`+sns.jointplot(x="x", y="y", data=df, kind="kde")`
	`20`	`+`
	`21`	`+f, ax = plt.subplots(figsize=(6, 6))`
	`22`	`+sns.kdeplot(df.x, df.y, ax=ax)`
	`23`	`+sns.rugplot(df.x, color="g", ax=ax)`
	`24`	`+sns.rugplot(df.y, vertical=True, ax=ax)`
	`25`	`+`
	`26`	`+f, ax = plt.subplots(figsize=(6, 6))`
	`27`	`+cmap = sns.cubehelix_palette(as_cmap=True, dark=0, light=1, reverse=True)`
	`28`	`+sns.kdeplot(df.x, df.y, cmap=cmap, n_levels=60, shade=True)`
	`29`	`+`
	`30`	`+g = sns.jointplot(x="x", y="y", data=df, kind="kde", color="m")`
	`31`	`+g.plot_joint(plt.scatter, c="w", s=30, linewidth=1, marker="+")`
	`32`	`+g.ax_joint.collections[0].set_alpha(0)`
	`33`	`+g.set_axis_labels("$X$", "$Y$")`
	`34`	`+`
	`35`	`+`
	`36`	`+`
	`37`	`+`
	`38`	`+`

`‎day-126/4.py`

Lines changed: 11 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,11 @@`
	`1`	`+import seaborn as sns`
	`2`	`+import matplotlib.pyplot as plt`
	`3`	`+from scipy import stats`
	`4`	`+sns.set(color_codes=True)`
	`5`	`+`
	`6`	`+iris = sns.load_dataset("iris")`
	`7`	`+sns.pairplot(iris)`
	`8`	`+`
	`9`	`+g = sns.PairGrid(iris)`
	`10`	`+g.map_diag(sns.kdeplot)`
	`11`	`+g.map_offdiag(sns.kdeplot, n_levels=6)`