Presentations / code snippets from PyData Kaunas meetups
Presentation: Movie Audience Score Calculation | Code: zaibacu/masters
Presenter: Šarūnas Navickas
About: My journey into making automatic Audience Score calculation system for movies using Lithuanian internet comments. Short summary and results of research done on this topic.
Presentation: Automated time series analysis, monitoring use case
Presenter: Egidijus Pilypas
About: Automated change and outlier detection in time series data, monitoring use case.
Presentation: Inception and Easy Transfer Learning | Code: dvisockas/retrain
Presenter: Danielius Visockas
About: Using to state-of-the art computer vision models (Inception and Easy Transfer Learning) for your own projects. There have been some huge recent advancements in the field of computer vision, but one of the biggest of them is the availability of the state-of-the-art models because of the ImageNet competition.
Presentation: Kaggle: A path to becoming an AI expert & winning data science competitions
Presenter: Darius Barušauskas
About: Kaggle is a platform for predictive modelling and analytics competitions in which companies and researchers post data and statisticians and data miners compete to produce the best models for predicting and describing the data.
Open Space
Presenter: Everyone
No presentation this time. Just open space for communication about Python and data.
Presentation: Scikit-learn: Machine Learning in Python
Presenter: Irina Matijosaitiene
Irina Matijosaitine is Data Science Researcher for Smart Cities, PhD. Assoc. Prof. Kaunas University of Technology/Yale University. Theme: data pre-processing ir classification tasks (logistic, Naive Bayes, Decision Tree, K-NN, svm). Based on Kaggle competitions.
Presentation: Building a Robot That Learns How to Drive by Himself
Presenter: Evaldas Kazlauskis
During 20 min presentation Evaldas will talk about how I built a simple self-driving robot using machine learning. Presentation will focus on ideas and methods behind artificial genetic evolution, how it resembles nature and how it can be used to solve tasks.
Workshop: Digit Recognition Workshop(tf/keras)
Presenter: Marius Ivaskevicius
The digit recognition workshop. What we'll do? Marius Ivaskevicius will lead a workshop based on https://www.kaggle.com/c/digit-recognizer kaggle competition.
Presentation: Online Machine Learning
Presenter: Darius Aliulis
Online Machine Learning deals with machine learning model updates. When new data arrives training new model every time is inefficient: online learning solves that. In this presentation we will cover Scikit-Learn models that allow PARTIAL FITTING. We will also be introduced to Vowpal Wabbit, a production-grade machine learning library for online learning. https://en.wikipedia.org/wiki/Online_machine_learning
Presentation: Analysis of Lithuanian texts: a case of moon and femininity
Presenter: Evaldas Vaičiukynas
Abstract: Presentation will discuss machine learning task of text classification. Text corpora was ASTRA stenograms, containing 110905 Lithuanian parliamentary transcripts from 147 speakers, collected during 1990 March - 2013 December. Texts were categorized by the political partisanship of a speaker, the gender of a speaker and the fact that a transcript was recorded around a full moon date. Types of pre-processing considered: original text, lemmized, morphized and translated to English. Lemmas and morphemes were obtained using semantika.lt and English translation using Google Translate services. Feature sets investigated: 6 from gensim (3 Doc2Vec variants, LSI, LDA, RP), 1 from fastText (Sent2Vec), and 3 custom-made (morfologija, stilometNER, ontologija). Random forest was used as a base-learner as well as a meta-learner (in 7 "stacking" configurations). Experiments reveal which categories, which types of pre-processing and which feature sets appear to be the most successful for texts analysed.