Thomas Dorfer

Data & Applied Scientist @Microsoft
Substack:

[Towards Data Science @TDataScience]

In machine learning, even the smallest data handling mistakes can derail your model's reliability.

[Towards Data Science @TDataScience]

Small, easily missed mistakes can really have a big impact on your model’s performance, particularly if they go unnoticed for an extended period of time. @ThomasADorfer highlights the pitfalls of relying on data that’s unavailable at prediction time and mixing magic numbers with real values.

https://t.co/90TbQWr5sv

[Towards Data Science @TDataScience]

Understanding XGBoost parameters is critical for building robust machine learning models. @ThomasADorfer dives into the nuances of parameters like max_depth, min_child_weight, and gamma, revealing how they balance complexity and performance.

https://t.co/lHVGX4ZwjP

[Towards Data Science @TDataScience]

. @ThomasADorfer offers strategies on how to keep up with a rapidly advancing field.
https://t.co/BaDAUatPKB

[Towards Data Science @TDataScience]

"Data visualization is an important aspect of data science that enables its practitioner to obtain a graphical representation of the data at hand, detect anomalies, or recognize patterns and trends."

Read more from @ThomasADorfer's post. https://t.co/NS73kLtFkB

[Towards Data Science @TDataScience]

. @ThomasADorfer offers a comprehensive overview of the remarkable progress that led AI to become a household name. https://t.co/VIqlkFNur0

[Towards Data Science @TDataScience]

. @ThomasADorfer revisits some of the key breakthroughs that got us to where we are today. https://t.co/VIqlkFNur0

[harpreet @DataScienceHarp]

An excellent piece about YOLO-NAS by ⁦⁦@ThomasADorfer⁩
https://t.co/1YIp0yIhdJ

[Towards Data Science @TDataScience]

Ten Years of AI in Review by @ThomasADorfer https://t.co/VIqlkFNur0

[Towards Data Science @TDataScience]

Reporting on the latest advances in object detection, @ThomasADorfer walks us through the inner workings of YOLO-NAS, a foundational model generated through neural architecture search, innovative quantization blocks, and a robust pre-training paradigm. https://t.co/GPhTqwBUPq

[Power BI Weekly @PowerBIWeekly]

Quick and Easy Time-Series Forecasting in #PowerBI: A Practical Guide from @ThomasADorfer https://t.co/okuGE3iH3q #PowerBI

[Thomas Dorfer @ThomasADorfer]

Exciting news for anyone interested in AI and finance! Two recent studies show that #ChatGPT is capable of deciphering Fedspeak (the deliberately vague statements made by the Federal Reserve) and predicting market moves based on corporate news headlines.
https://t.co/jqJJDSw0hX

[Towards Data Science @TDataScience]

Quick and Easy Time-Series Forecasting in PowerBI: A Practical Guide by @ThomasADorfer https://t.co/gOEYYbyWKP

[Thomas Dorfer @ThomasADorfer]

On April 22, the world came together to celebrate #EarthDay. To honor the occasion, I'd like to share a special edition of my newsletter Tech Talk with Thomas, focused entirely on how data science and AI can help solve sustainability problems.

https://t.co/KSq3LhBVNJ

[Towards Data Science @TDataScience]

How to Choose the Best Evaluation Metric for Regression Problems by @ThomasADorfer https://t.co/BNZyUocTnT

[Thomas Dorfer @ThomasADorfer]

@AllenDowney @TDataScience Thanks for the comment, Allen! In general, I agree with you. However, if that classifier is already in production or has been trained on a large dataset, it may not be very practical to just reverse the output. In that case, the line below the diagonal could still be useful.

[Towards Data Science @TDataScience]

Learn how to leverage AutoML to maximize the outcome of your machine learning workflows with @ThomasADorfer's guide. https://t.co/7VA9x662JP

[Towards Data Science @TDataScience]

. @ThomasADorfer offers a comprehensive guide covering the most commonly used evaluation metrics for supervised classification and their utility in different scenarios. https://t.co/SvhuxINAsB

[Thomas Dorfer @ThomasADorfer]

In this week's edition of Tech Talk with Thomas:

✂️ Meta’s Segment Anything Model (SAM)
🔗 Method chaining in Python

To find out more, check it out on Substack 👇

#datascience #ai #artificialintelligence #technology #python

https://t.co/OGVIXHf7g1…

[Towards Data Science @TDataScience]

. @ThomasADorfer offers a comprehensive guide that covers the most commonly used evaluation metrics for supervised classification and their utility in different scenarios. https://t.co/SvhuxIN2D3