Ian Ozsvald
where else to find them
what they talk about
all 11 in the filters
where they keep going back
- PyData London6 editions2016–2025
- PyData Global4 editions2020–2024
all 6 in the filters
on a stage with
who talks about the same things
- Vincent D Warmerdam22 talks
- Matt Harrison16 talks
- Giles Weaver5 talks
- Marysia Winkels9 talks
- Ramon Perez8 talks
- Reuven M. Lerner24 talks
| title | conference | date |
|---|---|---|
| Successful Projects through a bit of Rebellion | PyData London 2025 | 2025-06-06 |
Claims a private leadership community called Rebel AI peer mentorship group grew in 15 months and yields concrete tactics not fluff, shows a few multi‑million… | ||
| Valuable LLM lessons learnt on Kaggle's ARC AGI Challenge | PyData Global 2024 | 2024-12-03 |
Six months of work produces an automatic program generator that runs Llama3 locally via llama.cpp on an RTX3090, generates Python/NumPy solutions, executes… | ||
| Pandas 2, Dask or Polars? Tackling larger data on a single machine | PyData Global 2023 | 2023-12-06 |
Pandas 2 adds Arrow types and copy‑on‑write (to be default later) and on one dataset shows 82 million rows → 39 GB (numpy) vs 11 GB (Arrow) but mixed perf… | ||
| Data Science Project Patterns that Work | PyData Global 2022 | 2022-12-01 |
Practical patterns for the five project stages: choose good projects, de‑risk data by talking to users and using schemas and profilers, do productive research… | ||
| Building Successful Data Science Projects | PyData London 2022 | 2022-06-17 |
Clear specifics decide whether to watch: is there a written specification and de‑risking plan, and does the $1 million client claim trace to measurable… | ||
| Skinny Pandas Riding On A Rocket | PyData Global 2020 | 2020-11-11 |
25 million rows (4 GB CSV) inflate to ~10 GB in pandas memory; converting strings to categorical: 3.4 GB → 219 MB and downcasting numerics cuts RAM and speeds… | ||
| Making Pandas Fly | PyData Fest Amsterdam 2020 | 2020-06-15 |
Shows concrete wins: converting repeated strings to category dtype halves memory and cuts a value_counts from ~400ms to ~23ms, switching float64 to float32… | ||
| A gentle introduction to Pandas timeseries and Seaborn | PyData London 2019 | 2019-07-12 |
Uses Pandas and Seaborn in a Jupyter/Anaconda stack to manipulate a German utilities CSV (~4,383 rows over ~2006–2017) and shows concrete code: to_datetime… | ||
| Creating correct and capable classifiers | PyData Amsterdam 2018 | 2018-05-25 |
A practical workflow that starts with a concrete baseline and ends with explainable models the talk claims reduces guesswork and reveals real bugs: use… | ||
| Creating correct and capable classifiers | PyData London 2018 | 2018-04-26 |
Framework gives concrete checks: start with a dummy classifier baseline (Titanic example ≈0.62), then an untuned random forest reliably improves scores (≈0.83)… | ||
| Machine learning with ventilator data | PyData London 2017 | 2017-05-05 |
Processes 100Hz dual-channel ventilator data from neonatal ICU traces, about 160 ventilator days collected (each day ≈600–700 MB) producing roughly 43,000… | ||
| Using Machine Learning to solve a classification problem with scikit-learn | PyCon UK 2016 | 2016-09-15 |
Uses the Titanic dataset to show a reproducible Python workflow with scikit-learn, starting from a dummy classifier baseline (about 62% accuracy) and moving to… | ||
| Statistically Solving Sneezes and Sniffles Step by Step | PyData London 2016 | 2016-05-06 |
Project records a year of one-person, timestamped symptom logs with GPS and backs them up to Segment IO, plus external London Air, weather, MyFitnessPal and… | ||
| Statistically Speculating on the Source of Sneezes and Sniffles | PyData London 2016 | 2016-05-06 |
One year of self‑reported antihistamine usage plus sneeze counts and GPS traces feed a homemade iPhone logging app and HDF5/pandas pipeline; external features… | ||
| Experiences making CPU-bound tasks run much faster | EuroPython 2011 | 2011-06-20 |
Profiling with cProfile and line_profiler finds hot loops, then applied PyPy, NumPy vectorization, numexpr, C extensions and GPU kernels to the hot spots. One… | ||
| nothing matches | ||