PyData Berlin 2014
2014-07-25..2014-07-27
Berlin, Germany
[PyData Berlin 2014, PyData Berlin 2015, PyData Berlin 2016, PyData Berlin 2017, PyData Berlin 2018]
| title | author | length |
|---|---|---|
| Lightning Talks | 36m | |
Quick-fire demos and community updates give concrete claims and tools: building a PyData track in Florence (plans and dates), video analytics with NumPy OpenCV… | ||
| Pandas' Thumb: unexpected evolutionary use of a Python library. | Chris Nyland | 35m |
Claim: Pandas can replace spreadsheets for complex UK SDLT lease work by building time‑aware data frames that do the arithmetic, handle NPVs and credits, and… | ||
| Massively Parallel Processing with Procedural Python | Ronert Obst | 41m |
Claims that Python runs inside the database to avoid moving terabytes of data to memory, using PL/Python to call numpy scikit-learn and NLTK on nodes and… | ||
| IPython and Sympy to Develop a Kalman Filter for Multisensor Data Fusion | Paul Balzer | 47m |
Clear claim: a working Kalman filter for a nonlinear vehicle model that fuses acceleration, speed and GPS by deriving analytic matrices symbolically with… | ||
| Intro to ConvNets | Kashif Rasul | 44m |
A practical introduction to convolutional nets that gives concrete claims and commands not platitudes: explains forward/backprop and regularisation, shows… | ||
| Make sense of your (big) data using Elasticsearch | Honza Kral | 47m |
Shows concrete claims and demos: Elasticsearch indexes real JSON documents, runs full‑text and structured queries, faceting and nested/parent‑child queries… | ||
| The challenges and frontiers of data science in Europe | 01h 00m | |
Panel discussion reports concrete wins and failures: a hedge‑fund manager wires up an numpy convolution experiment to test a new signal, Royal Mail commits to… | ||
| Conda: a cross-platform package manager for any binary distribution | Ilan Schnell | 45m |
Conda claims to solve Python packaging by treating Python and C libraries the same, shipping only binaries with metadata in an index and using a solver that… | ||
| Exploring Patent Data with Python | Franta Polach | 41m |
A patent recommendation engine for search uses Pythonic NLP and topic modeling tools Gensim + NLTK + LDA to index US patents (1975–present) totaling 70 GB… | ||
| Building the PyData Community | Travis E. Oliphant | 44m |
Describes building PyData around a nonprofit called num Focus claiming US 501c3 status to fund numpy scipy pandas IPython and community activities, funded… | ||
| Faster than Google? Optimization lessons in Python. | Radim Řehůřek | 28m |
Porting word2vec into Python produces concrete results: an optimized Python port using generators, NumPy, Cython, BLAS runs substantially faster than naive… | ||
| Introduction to the Signal Processing and Classification Environment pySPACE | Mario Michael Krell | 41m |
pySPACE claims to let users specify complete signal‑processing and classification pipelines in YAML and run large automated benchmarks in parallel, with over… | ||
| ABBY: A Django app to document your A/B tests | Andy Goldschmidt | 33m |
ABBY is a Django app that centralizes A/B test records for a company of about 180 people at Jimdo and is now open sourced on GitHub. It provides a RESTful API… | ||
| Commodity Machine Learning | Andreas Mueller | 43m |
Argues machine learning is becoming a commodity by showing concrete wins: psyched learn plus pandas and IPython Notebook let common supervised tasks be solved… | ||
| Speed Without Drag | Saul Diez-Guerra | 40m |
Practical walkthrough evaluates concrete gains across low‑friction numerical Python tools using a generated diagonal polygon and the shoelace formula as the… | ||
| Python and pandas as back end to real-time data driven applications | Giovanni Lanzani | 40m |
Built a web app using Python and pandas with a Flask + AngularJS front end, Hadoop for raw storage and postgress for serving. Logs shrink from 40 terabytes to… | ||
| How to Spy with Python | Lynn Root | 29m |
Covers specific claims and concrete tools: reproduces PRISM and X key score–style queries and shows scale claims (hundreds of servers, Utah data center). Walks… | ||
| Algorithmic Trading with Zipline | Thomas Wiecki | 38m |
Zipline is a BSD-licensed, event-driven backtester that walks historical data row by row and prevents look‑ahead bias while outputting results as Pandas… | ||
| Low-rank matrix approximations in Python | Christian Thurau | 36m |
Concrete claims: a Python toolbox called PyMF implements PCA, k‑means, NMF and archetypal analysis and can operate on disk‑backed arrays via HDF5. The talk… | ||
| Street Fighting Trend Research | Benedikt Koehler | 48m |
Hands-on recipes show how to scrape and analyze research pages, conference proposals and Foursquare venue graphs using Pandas and NTLK inside an IPython… | ||
| Parallel processing using python and gearman | Pedro Miguel Dias Cardoso | 34m |
Processes load large NLP models so processes cannot be spawned per item, models reach 0.5 GB and Synthesio handles 20–40 million verbatims per day. Solution… | ||
| Data Oriented Programming | Francesc Alted | 36m |
Claim: CPU cycles are cheap but memory access is the bottleneck, so optimizing data layout beats naive parallelism. Examples and numbers: numpy 1.6s ->… | ||
| Interactive Analysis of (Large) Financial Data Sets | Yves J. Hilpisch | 45m |
Shows concrete results: uses NumPy, pandas, PyTables/HDF5, IPython/Jupyter and multiprocessing to do portfolio optimization, PCA and Monte Carlo; runs 2,500… | ||
| Mall Analytics Using Telco Data & Pandas | Karolina Alexiou | 28m |
Estimates mall visitor trends from telco event records using Hadoop/map‑reduce, HDFS, Avro and Python, producing CSVs for a dashboard and preserving privacy… | ||
| Driving Moore's Law with Python-Powered Machine Learning: An Insider's Perspective | Trent McConaghy | 46m |
Claims that machine learning keeps Moore’s Law moving by absorbing manufacturing variation into models and CAD tools, with concrete stacks: Python 2.7 with… | ||
| Semantic Python: Mastering Linked Data with Python | Valerio Maggio | 48m |
Claims that rdflib lets Python code create, query and persist RDF graphs, demonstrates rdflib + NetworkX integration by crawling FOAF data, running SPARQL… | ||
| Quantified Self: Analyzing the Big Data of our Daily Life | Andreas Schreiber | 44m |
Claims that self-tracking yields actionable results — blood pressure logs lead to medication changes — and demonstrates the actual tooling and limits: pandas… | ||
| Generators Will Free Your Mind | James Powell | 49m |
Claims that generators remove API-mode assumptions and avoid needless work with lazy streams. Demos show a 50-line meeting scheduler, a trading-query case… | ||
| Packaging and Deployment | Travis E. Oliphant | 01h 14m |
Shows concrete, repeatable packaging: use conda environments to install Python, numpy, scipy and switch between 2.7/3.x; conda uses a SAT solver and a simple… | ||
| Visualising Data through Pandas | Vincent D Warmerdam | 01h 04m |
Demo shows pandas data frame as the main in‑memory workhorse, using IPython notebooks to load CSVs, resample time series, and run groupby split apply combine… | ||
| Dealing With Complexity | Jean-Paul Schmetz | 42m |
Defines complexity as whatever prevents doing the right thing now or later and argues simplicity is the prerequisite for agility, reliability and scale. Gives… | ||
| Extract Transform Load using mETL | Bence Faludi | 01h 04m |
mETL is a Python ETL package built for Central European University that loads CSV/JSON/XML into MySQL or PostgreSQL, runs field-by-field transformations in… | ||
| nothing matches | ||