Sports Analytics Weekly by kubeia.io - 14/2023
- Week 52
- Week 51
- Week 49
- Week 48
- Week 47
- Week 46
- Week 45
- Week 44
- Week 43
- Week 42
- Week 41
- Week 40
- Week 39
- Week 38
- Week 37
- Week 36
- Week 35
- Week 34
- Week 33
- Week 32
- Week 31
- Week 30
- Week 29
- Week 28
- Week 27
- Week 26
- Week 25
- Week 24
- Week 23
- Week 22
- Week 21
- Week 20
- Week 19
- Week 18
- Week 16
- Week 14
- Week 13
- Week 12
- Week 10
- Week 9
- Week 6
- Week 5
- Week 4
- Week 3
- Week 1
🎲 Betting
Football Betting for Beginners
bettinganalytics.io
In Football Betting for Beginners, Antoine J.W. Martin supplies, in a form suitable for laymen, guidance in the creation of data-driven rating systems that seek to identify value in the bookmakers’ odds.
âš˝ Sports
Come In Number 5: Line-Breaking Midfielders In The Premier League
statsbomb.com
Last year I used StatsBomb 360 data to take a look at players breaking lines with passes and with prizes down at “Chez StatsBomb” extending to “write more articles” this year I figured to take another look and see what jumped out. And then as we will see, I spied a hot young talent and investigated them.
Champions League Betting Blueprint: Quarter Finals preview (2022/23)/ Pinnacle Podcast
apple.com
The Champions League betting blueprint returns to preview the 2022/23 quarter finals which features holders Real Madrid, favourites Manchester City and Serie A runaway leaders Napoli. The team of Gareth Wheeler, Jake Osgathorpe and Andrew Beasley take a look at Pinnacle's outright markets, match odds and help our bettors find the best value plays.
đź’° Quantitative Finance
BloombergGPT: A Large Language Model for Finance
arxiv.org
The use of NLP in the realm of financial technology is broad and complex, with applications ranging from sentiment analysis and named entity recognition to question answering. Large Language Models (LLMs) have been shown to be effective on a variety of tasks; however, no LLM specialized for the financial domain has been reported in literature. In this work, we present BloombergGPT, a 50 billion parameter language model that is trained on a wide range of financial data. We construct a 363 billion token dataset based on Bloomberg's extensive data sources, perhaps the largest domain-specific dataset yet, augmented with 345 billion tokens from general purpose datasets. We validate BloombergGPT on standard LLM benchmarks, open financial benchmarks, and a suite of internal benchmarks that most accurately reflect our intended usage. Our mixed dataset training leads to a model that outperforms existing models on financial tasks by significant margins without sacrificing performance on general LLM benchmarks. Additionally, we explain our modeling choices, training process, and evaluation methodology. As a next step, we plan to release training logs (Chronicles) detailing our experience in training BloombergGPT.
pandas 2.0 and the Arrow revolution
datapythonista.me
At the time of writing this post, we are in the process of releasing pandas 2.0. The project has a large number of users, and it's used in production quite widely by personal and corporate users. This large use based forces us to be conservative and make us avoid most big changes that would break existing pandas code, or would change what users already know about pandas. So, most changes to pandas, while they are important, they are quite subtle. Most of our changes are bug fixes, code improvements and clean up, performance improvements, keep up to date with our dependencies, small changes that make the API more consistent, etc. A recent change that may seem subtle and it's easy to not be noticed, but it's actually very important is the new Apache Arrow backend for pandas data.
🤖 Machine Learning
Large Language Models 3.0
sebastianraschka.com - Sebastian Raschka
It's been about half a decade since we saw the emergence of the original transformer model, BERT, BLOOM, GPT 1 to 3, and many more.