mirror of
https://github.com/Estom/notes.git
synced 2026-02-07 04:23:55 +08:00
71 lines
2.0 KiB
Markdown
71 lines
2.0 KiB
Markdown
# 贝叶斯更新
|
|
|
|
此动画显示在新数据到达时重新安装的后验估计更新。
|
|
|
|
垂直线表示绘制的分布应该收敛的理论值。
|
|
|
|
|
|
|
|
```python
|
|
import math
|
|
|
|
import numpy as np
|
|
import matplotlib.pyplot as plt
|
|
from matplotlib.animation import FuncAnimation
|
|
|
|
|
|
def beta_pdf(x, a, b):
|
|
return (x**(a-1) * (1-x)**(b-1) * math.gamma(a + b)
|
|
/ (math.gamma(a) * math.gamma(b)))
|
|
|
|
|
|
class UpdateDist(object):
|
|
def __init__(self, ax, prob=0.5):
|
|
self.success = 0
|
|
self.prob = prob
|
|
self.line, = ax.plot([], [], 'k-')
|
|
self.x = np.linspace(0, 1, 200)
|
|
self.ax = ax
|
|
|
|
# Set up plot parameters
|
|
self.ax.set_xlim(0, 1)
|
|
self.ax.set_ylim(0, 15)
|
|
self.ax.grid(True)
|
|
|
|
# This vertical line represents the theoretical value, to
|
|
# which the plotted distribution should converge.
|
|
self.ax.axvline(prob, linestyle='--', color='black')
|
|
|
|
def init(self):
|
|
self.success = 0
|
|
self.line.set_data([], [])
|
|
return self.line,
|
|
|
|
def __call__(self, i):
|
|
# This way the plot can continuously run and we just keep
|
|
# watching new realizations of the process
|
|
if i == 0:
|
|
return self.init()
|
|
|
|
# Choose success based on exceed a threshold with a uniform pick
|
|
if np.random.rand(1,) < self.prob:
|
|
self.success += 1
|
|
y = beta_pdf(self.x, self.success + 1, (i - self.success) + 1)
|
|
self.line.set_data(self.x, y)
|
|
return self.line,
|
|
|
|
# Fixing random state for reproducibility
|
|
np.random.seed(19680801)
|
|
|
|
|
|
fig, ax = plt.subplots()
|
|
ud = UpdateDist(ax, prob=0.7)
|
|
anim = FuncAnimation(fig, ud, frames=np.arange(100), init_func=ud.init,
|
|
interval=100, blit=True)
|
|
plt.show()
|
|
```
|
|
|
|
## 下载这个示例
|
|
|
|
- [下载python源码: bayes_update.py](https://matplotlib.org/_downloads/bayes_update.py)
|
|
- [下载Jupyter notebook: bayes_update.ipynb](https://matplotlib.org/_downloads/bayes_update.ipynb) |