sengbao_mle
翻新的最大似然估计框架
copied and renewed maximum likelihood estimation package from python-mle
搬运与翻新说明
- 本仓库搬运自ibab/python-mle
- 由于旧仓库最后一次更新于2016年2月19日,其中一些结构安排和依赖库已经发生了变化,因此笔者将其翻新后,重新发布到pypi上,便于直接安装调用
安装&使用指南🎯
- 安装
使用
pip install sengbao_mle命令进行安装
- 使用
- 导入框架
import sengbao_mle as m
- 使用实例1——原库示例(翻新后)
import numpy as np import sengbao_mle as m # Define model x = m.var('x', observed=True, vector=True) y = m.var('y', observed=True, vector=True) a = m.var('a') b = m.var('b') sigma = m.var('sigma') model = m.Normal(y, a * x + b, sigma) # Generate data xs = np.linspace(0, 2, 20) ys = 0.5 * xs + 0.3 + np.random.normal(0, 0.1, 20) # Fit model to data result = model.fit({'x': xs, 'y': ys}, {'a': 1, 'b': 1, 'sigma': 1}) print(result)
- 使用实例2——自定义分布
import numpy as np import sengbao_mle as m import theano.tensor as T # 以Subbotin分布为例 class Subbotin(m.Model): def __init__(self,x,af,*args,**kwargs): # 自定义的分布类型,均需继承自m.Model super(Subbotin, self).__init__(*args,**kwargs) # 运算都需要采用theano.tensor中的运算函数 mm=T.mean(x) xg=T.mean(T.abs_(x-mm)**af)**(1/af) # 写入该分布的对数概率密度函数 self._logp=m.distributions.bound(T.log( T.exp(-((T.abs_((x-mm)/xg))**af)/af)/(2*xg*(af**(1/af))*T.gamma(1+1/af)) )) # 添加样本点 self._add_expr('x',x) # 添加要拟合的参数 self._add_expr('af',af) # 样本点 x = m.var('x', observed=True, vector=True) # 目标参数 af=m.var('af') model=Subbotin(x,af) xs=np.random.normal(-3,3,(1000,)) # 给目标参数设置初始值 result = model.fit({'x': xs}, {'af': 1}) print(result)
- 使用实例3——遍历初始值
# 由于目标参数的初始值的设定,对能否拟合成功有着较大影响,因此在拟合失败时,可以考虑便利初始值,以寻找能拟合成功的初始参数 import numpy as np import sengbao_mle as m import theano.tensor as T # 以Subbotin分布为例 class Subbotin(m.Model): def __init__(self,x,af,*args,**kwargs): # 自定义的分布类型,均需继承自m.Model super(Subbotin, self).__init__(*args,**kwargs) # 运算都需要采用theano.tensor中的运算函数 mm=T.mean(x) xg=T.mean(T.abs_(x-mm)**af)**(1/af) # 写入该分布的对数概率密度函数 self._logp=m.distributions.bound(T.log( T.exp(-((T.abs_((x-mm)/xg))**af)/af)/(2*xg*(af**(1/af))*T.gamma(1+1/af)) )) # 添加样本点 self._add_expr('x',x) # 添加要拟合的参数 self._add_expr('af',af) # 样本点 x = m.var('x', observed=True, vector=True) # 目标参数 af=m.var('af') model=Subbotin(x,af) xs=np.random.normal(-3,3,(1000,)) # 遍历能拟合成功的初始参数值 rs=[] for af in range(-100,100): result = model.fit({'x': xs}, {'af': af}) if result.success is True: rs.append({af:result}) af=np.mean([i.values().x['af'] for i in rs]) print(af)
相关链接🔗
Metadata
Release files for sengbao-mle 0.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| sengbao_mle-0.0.1.tar.gz | 10.3 kB | Without provenance | Details for sengbao_mle-0.0.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| sengbao_mle-0.0.1-py3-none-any.whl | Python 3 | none | any | Details for sengbao_mle-0.0.1-py3-none-any.whl |
Total release size: 20.9 kB
Release files / sengbao_mle-0.0.1.tar.gz
| Download URL | sengbao_mle-0.0.1.tar.gz |
|---|---|
| Size | 10.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
1ee5f5aac0ff5fd52bf1054d2d0b8bab4871622906133e80b85e97b6e8480cef
|
|
BLAKE2b-256 checksum How to use checksums |
ae38207531c36993ebc65c5b1e3547c66f95dfa20b7fda80901d91771844e57c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.1 CPython/3.9.15
|
Release files / sengbao_mle-0.0.1-py3-none-any.whl
| Download URL | sengbao_mle-0.0.1-py3-none-any.whl |
|---|---|
| Size | 10.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
faf5d34e5a4ff7c81201fe04617fa4c4b594955cbca960b1a289e0f01f631de2
|
|
BLAKE2b-256 checksum How to use checksums |
72faecc146c03790372789be5a3ec4e8e3347605c1d525d5af956a99b2e5189e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.1 CPython/3.9.15
|
Security for sengbao-mle 0.0.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.