pyro-ppl/pyro

[FR] get mcmc.summary() as DataFrame

Open

#3,013 创建于 2022年1月26日

在 GitHub 查看
 (6 评论) (1 反应) (0 负责人)Python (8,211 star) (981 fork)batch import
enhancementhelp wantedusability

描述

First of all thanks for the great package!

Issue Description

One of the useful features of the MCMC class is the MCMC.summary() method which prints summary statistics of the sampled chain.

In this issue I'm asking for the feature to be able to get this table as a Pandas DataFrame.

Code Snippet

def model(data):
    ...

nuts_kernel = NUTS(model)
mcmc = MCMC(nuts_kernel, num_samples=500)
mcmc.run(data)
mcmc.summary()

It would be very helpful if there was a simple way to get mcmc.summary as DataFrame or perhaps alternatively as a dict which can become a DataFrame.

贡献者指南

技术栈
pythonpandas
领域
datamachine learning
议题类型
feature
难度面向新贡献者的预计实现难度，1 表示很小改动，5 表示专家级工作。
2
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
1-3 hours
活动状态议题当前的可参与程度：新鲜、活跃、陈旧、阻塞或等待维护者输入。
fresh
清晰度议题是否清楚说明期望改动、验收标准和下一步。
clear
前置要求
familiarity with Pyro's MCMC classbasic pandas knowledge
新手友好度1-100 的估计分数，表示该议题对首次贡献者的友好程度。
75
研究方向
The feature request is to extend the `MCMC.summary()` method (likely defined in `pyro/infer/mcmc.py`) to optionally return a pandas DataFrame. The current implementation prints a table. A new parameter (e.g., `as dataframe=False`) could be added to return a DataFrame when set to True. The summary statistics are stored internally; extracting them into a dict and converting to DataFrame should be straightforward. Check the existing ` summary` helper and the `Diagnostics` module for formatting. No prior PRs exist for this issue.