Ian Ozsvald

data_science_delivered. Observations from Ian on successfully delivering data science products

ipython_memory_usage. IPython tool to report memory usage deltas for every command you type

EuroPython2011_HighPerformanceComputing. Code for High Performance Computing tutorial for EuroPython 2011

twitter-text-python. Twitter text processing library (auto linking and extraction of usernames, lists and hashtags). Based on the Java implementation by Matt Sanford

dtype_diet. Tries to shrink your Pandas column dtypes with no data loss so you have more spare RAM

featherweight_web_api. Featherweight web API provider for serving R&D methods as web functions

ark-tweet-nlp-python. Simple Python wrapper around runTagger.sh of ark-tweet-nlp

twitter-social-graph-networkx. Download, summarise and visualise the followers in a small twitter social network

social_media_brand_disambiguator. Brand disambiguator for tweets to differentiate e.g. Orange vs orange (brand vs foodstuff), using NLTK and scikit-learn

HighPerformancePython_PyCon2012. Python

beyond_correlation. Exploratory code to see if we can learn about feature relationships in a DataFrame using machine learning

notes_to_self. Python

twitter_networkx_concept_map. Take streaming tweets, extract hashtags & usernames, create graph, export graphml for Gephi visualisation

string_distance_metrics. A set of Python string distance metrics for string distance comparisons

learning_text_transformer_demo. Demo code for learning_text_transformer

pyconuk_using_sklearn_classification. PyConUK 2016 talk - Using Machine Learning to solve a classification problem with scikit-learn - a practical walkthrough

pycon2013_applied_parallel_computing. Applied Parallel Computing tutorial material for PyCon 2013 (Minesh Amin, Ian Ozsvald)

learning_text_transformer. Search 'from' and 'to' strings to learn a text cleaning mapping

python_template_with_config. Python coding template including ENV environment variable configuration

euroscipy2014_highperformancepython. Python

mot_pandas2_polars_dask. Investigation for PyDataLondon 2023 and ODSC 2023 conference comparing Pandas 2, Polars and Dask

ParallelPython_EuroSciPy2012. Starting code & solutions for EuroSciPy Paralllel Python 2 hour tutorial

Python6hrsTutorial. 6 hour Python tutorial slides (open sourced for anyone to re-use)

callgraph_james_powell. Code from James for a callgraph monitor

Mandelbrot_pyCUDA_Cython_Numpy. Compare execution speeds of Mandelbrot using pyCUDA, Cython, Numpy (with Didrik Pinte @ Enthought)

python-twitter. Python

plaquereader. Automatic English Heritage plaque transcriber for the OpenPlaques project

morris_counter. Probabilistic Morris Counter (counts 2^n using e.g. just a byte)

PyDataBudapest202004. "Making Pandas Fly" talk on data science at PyDataBudapest

research_module_layout_template. Python module structure for Jupyter Notebooks for R&D (e.g. Machine Learning) and collaboration

london_oyster_pdf_to_dataframe_parser. Convert London Oytser rail+tube+bus PDF journey histories into a Pandas DataFrame as an HDF5 file

talkpay_tweet_visualisataion. Viz of #talkpay pay tweets during May 2015

kaggle_titanic_ipythonnotebook_boilerplate. Boilerplate example using IPython Notebook to solve simplest (sex-field only) Titanic challenge for Kaggle (this will get you started wtih the Kaggle competition)

dataframe_visualiser. Make simple high-level visualisations of a Pandas dataframe to help when starting new data projects

SocialMicroprinter. CBM 231 drivers for Aruino+WiShield to turn a receipt printer into a remote social microprinter

oyster-scraper. Scrape the London Oyster website for Tube travel data with Python (forked from markk)

pylondinium2019. Jupyter Notebook

ocr_aicookbook. Optical Character Recognition server behind bottle.py interface

pydata2020global. Conference talk material

playgroup_llm_202506. Ian's playgroup deep dive on LLMs day for 2025-06

matplotlib_heightmap_2d_3d. Plot a heightmap (Z on X/Y axis) in 3D and 2D in matplotlib

uplift_experiment. Jupyter Notebook

sklearn_tutorial. Materials for my scikit-learn tutorial

tutorials-scikit-learn. Scikit-Learn tutorials

churn_a_b_tests_statistical_power. Jupyter Notebook

dtype_pandas_numpy_speed_test. Speed test on Pandas or NumPy

explore_cpu_cache_effect. Understanding my CPU's L3 cache using numpy vectorized operations

bulwark. Bulwark is a package for convenient property-based testing of pandas dataframes, supported for Python 3.5+.

pyconby2021. Material for PyCon Belarus 2021

dabl_experiments. Light explorations with dabl

playgroup_llm_202509. Python

quantified_cat_flap. RaspberryPi code to read reed switches on a catflap

pandas. Flexible and powerful data analysis / manipulation library for Python, providing labeled data structures similar to R data.frame objects, statistical functions, and much more

matplotlib_templates. Templates for things that I like