Data Structures for Statistical Computing in Python

Data Structures for Statistical Computing in Python

View more presentations from wesm

Bayesian estimation with Markov Chain Monte Carlo

A Gentle Introduction to Bayesian Biostatistics

Markov Chain Monte Carlo Tutorial

R Installer

NumPy

PyMC

MapReduce Algorithm In Python

Google's MapReduce in 98 Lines of Python

MapReduce Example Programs

Writing An Hadoop MapReduce Program In Python

MapReduce: Simplified Data Processing on Large Clusters

Octopy

Cluster Computing and MapReduce Lecture 1

Understanding Python

Guido van Rossum - Author of Python

Interview With Guido van Rossum

Part I: The Making of Python

Part II: Python's Design Goals

Part III: Programming at Python Speed

Part IV: Contracts in Python

Part V: Strong versus Weak Typing

Part VI:Designing with the Python Community

Towards a Standard Parser Generator

Abstract

Developing parsers for "little" languages is a common task for many software developers. People have frequently requested inclusion of a specific parser generator framework into the Python library. In this paper, we compare several Python parser generators, using the XPath language as an application. Criteria for comparison are ease of use, performance, and ease of deployment. Using these criteria, we give a recommendation for including parser generators into the standard library..

Keywords: Parsers, XPath, YAPPS, Spark, BisonGen

continue...

Algorithm Education in Python

Abstract

Design and analysis of algorithms are a fundamental topic in computer science and engineering education. Many algorithms courses include programming assignments to help students better understand the algorithms. Unfortunately, the use of traditional programming languages forces students to deal with details of data structures and supporting routines, rather than algorithm design. Python represents an algorithm-oriented language that has been sorely needed in education. The advantages of Python include its textbook-like syntax and interactivity that encourages experimentation. More importantly, we report our novel use of Python for representing aggregate data structures such as graphs and flow networks in a concise textual form, which not only encourages students to experiment with the algorithms but also dramatically cuts development time. These features have been implemented in a graduate level algorithms course with successful results.

continue...