blackjack-basic-strategy VS open-data

It’s an easy to use inference server for computer vision models.

The end result is a Docker container that serves a standardized API as a microservice that your application uses to get predictions from computer vision models (though there is also a native Python interface).

It’s backed by a bunch of component pieces:

* a server (so you don’t have to reimplement things like image processing & prediction visualization on every project)

* standardized APIs for computer vision tasks (so switching out the model weights and architecture can be done independently of your application code)

* model architecture implementations (which implement the tensor parsing glue between images & predictions) for supervised models that you've fine-tuned to perform custom tasks

* foundation model implementations (like CLIP & SAM) that tend to chain well with fine-tuned models

* reusable utils to make adding support for new models easier

* a model registry (so your code can be independent from your model weights & you don't have to re-build and re-deploy every time you want to iterate on your model weights)

* data management integrations (so you can collect more images of edge cases to improve your dataset & model the more it sees in the wild)

* ecosystem (there are tens of thousands of fine-tuned models shared by users that you can use off the shelf via Roboflow Universe[1])

Additionally, since it's focused specifically on computer vision, it has specific CV-focused features (like direct camera stream input) and makes some different tradeoffs than other more general ML solutions (namely, optimized for small-fast models that run at the edge & need support for running on many different devices like NVIDIA Jetsons and Raspberry Pis in addition to beefy cloud servers).

[1] https://universe.roboflow.com

* Most of the time I find Roboflow extremely handy, I used it to merge datasets, augmentate, read tutorials and that kind of thing. Basically you just create your dataset with roboflow and focus on other aspects.

For computer vision, there are 100k+ open source classification, object detection, and segmentation datasets available on Roboflow Universe: https://universe.roboflow.com

Solid website and app overall for learning more about computer vision, discovering datasets, and keeping up with advancements in the field: * https://roboflow.com/learn * https://universe.roboflow.com (datasets) | https://blog.roboflow.com/computer-vision-datasets-and-apis/ * https://blog.roboflow.com

If you're lacking training images, you can also use [Roboflow Universe](https://universe.roboflow.com) to obtain them (over 100 million labeled images available)

Roboflow | Multiple Roles | Full-time (Remote) | https://roboflow.com/careers

Roboflow is the fastest way to use computer vision in production. We help developers give their software the sense of sight. Our end-to-end platform[1] provides tooling for image collection, annotation, dataset exploration and curation, training, and deployment.

Over 100k engineers (including engineers from 2/3 Fortune 100 companies) build with Roboflow. And we now host the largest collection[2] of open source computer vision datasets and pre-trained models[3].

We have several openings available, but are primarily looking for strong technical generalists who want to help us democratize computer vision and like to wear many hats and have an outsized impact. (We especially love hiring past and future founders.)

We're hiring 3 full-stack engineers this quarter and we're also looking for an infrastructure engineer with Elasticsearch experience.

[1]: https://docs.roboflow.com

[2]: https://blog.roboflow.com/computer-vision-datasets-and-apis/

[3]: https://universe.roboflow.com

Based on what I do/use when I prepare models: A good framework for creating and improving this dataset faster is to use Roboflow Universe and search “flowers” and “bouquets of flowers” in the search bar (it’s like Google Images for CV Datasets). You can search images by subject, or metadata, and clone them directly into a free public workspace (they house up to 10k images without charge). * https://universe.roboflow.com/ * https://universe.roboflow.com/search?q=flowers * https://universe.roboflow.com/search?q=bouqets

Lots of ideas will come to mind if you look and search through open source datasets: https://universe.roboflow.com/

Open source datasets: https://universe.roboflow.com/ Model training: https://docs.roboflow.com/train Model deployment: https://docs.roboflow.com/inference/hosted-api

You can check statsbomb open data but you will to preprocess it from json to sql. They have great course and articles about analyzing the data. Another good reading is awasome-football . They provide list of resources to get data. But the most comprehensive and recommended resources eddwebster's guide. He worked for city football group and his repository updated frequently.

I tried visualising Enzo's progressive passes in each of his world cup matches. I used the data available on StatsBomb for this.

https://www.football-data.org/coverage https://datahub.io/collections/football https://github.com/statsbomb/open-data https://www.kaggle.com/datasets/hugomathien/soccer https://www.kaggle.com/datasets/martj42/international-football-results-from-1872-to-2017 https://www.kaggle.com/datasets/secareanualin/football-events https://www.kaggle.com/datasets/adityadesai13/european-football-database-20192020 https://www.kaggle.com/datasets/vivovinco/20212022-football-player-stats https://www.kaggle.com/datasets/antoinekrajnc/soccer-players-statistics

StatsBomb | Multiple roles | REMOTE, or Bath (UK), or Cairo (Egypt)

StatsBomb is a sports analytics startup, covering football (both the soccer and American varieties) and soon basketball. We sell data products as well as analysis tools to sports, media and gambling organisations, with a tech pipeline that includes computer vision, machine learning, stream processing, and web-based dataviz. We count many of the biggest names in football as customers, and your work will have a direct impact on our ability to deliver insights to those customers, driving success on the field.

We're hiring software engineers of various stripes (data pipeline roles with Python and Clojure, full-stack web dev roles with JavaScript) and more besides. We're fully remote, but have offices in Bath, UK and Cairo, Egypt for those that want them. We organise regular team days and also run our own industry-leading conference each year.

- Apply at: https://statsbomb.com/careers

If you'd like to find out more about football analytics:

- Play with our open data: https://github.com/statsbomb/open-data

- Read our articles: https://statsbomb.com/articles/

- Browse our conference videos: https://www.youtube.com/channel/UCmZ2ArreL9muPvH49Gaw0Bw

The data source providing all the passes is from StatBomb

If you're familiar with GitHub and JSON then https://github.com/statsbomb/open-data looks decent.

The company StatsBomb, which specializes in football analytics, has made a lot of their data available for public use here: https://github.com/statsbomb/open-data I’ve been playing with it recently and I’ve found it to be pretty great.

Messi has routinely been one of the best performers in European soccer, including his record-breaking 2011-2012 season in the Spanish league (“La Liga”) with Barcelona, where he set the record for most goals in a season. Unfortunately, success with the Argentina national team has frequently eluded him, finishing as a “runner-up” in the World Cup once and in the Copa America 3 times, before finally winning the Copa America in 2021. Critics often point to his difficulties with his national team as a fatal flaw. I was interested in how his scoring opportunities during arguably his best performance at Barcelona compared to his chances made with Argentina. The data suggests that he is regularly shooting from further away from goal when playing with Argentina when compared to his best performance with Barcelona, which could be a result of a number of factors (different team tactics, difficulty getting up the field, increasing age, less familiarity with teammates, etc.). Data: 2011/2012 La Liga season and World Cup 2018 data were collected from the very nice, public datasets provided by StatsBomb at https://github.com/statsbomb/open-data. The World Cup 2014 data was a bit more difficult to find, but was scraped from the Huffington Post . The StatsBomb data has a ton of great stats to dig into, but because the Huffington Post data had less detail, I wasn't able to go into all of it with just this plot.

I think Statsbomb has a free API you can use on Github if you request access. https://github.com/statsbomb/open-data