| searchable |
Delighted that a project that took shape within a #machinelearning
workshop at the @eScienceCenter and which was initiated and led by #RozaKamioglu
and #DisaSauter resulted in a data analysis of human #laughter sounds.
--> https://royalsocietypublishing.org/doi/10.1098/rsbl.2024.0543#d1e1021
I continue to work on #opensource versions of my various teaching materials. Here is the first complete draft for my Python Introduction (for the moment only in German, but English version is on the Todo-list).
New version of my (renamed) textbook: "Hands-on Introduction to Data Science using Python"
Content is now mostly complete. Text and figures will undergo further polishing.
--> https://florian-huber.github.io/data_science_course/book/cover.html
Nice collaboration led by Niek de Jonge just got published in Journal of Cheminformatics 🚀.
In this work, we implemented and evaluated an extensive cleaning pipeline for MS/MS data.
https://link.springer.com/article/10.1186/s13321-024-00878-1
#Python #matchms #cheminformatics #opensource #openscience
Big thanks to all co-authors for this very nice collaboration!
Mass spectral libraries have proven to be essential for mass spectrum annotation, both for library matching and training new machine learning algorithms. A key step in training machine learning models is the availability of high-quality training data. Public libraries of mass spectrometry data that are open to user submission often suffer from limited metadata curation and harmonization. The resulting variability in data quality makes training of machine learning models challenging. Here we present a library cleaning pipeline designed for cleaning tandem mass spectrometry library data. The pipeline is designed with ease of use, flexibility, and reproducibility as leading principles.Scientific contributionThis pipeline will result in cleaner public mass spectral libraries that will improve library searching and the quality of machine-learning training datasets in mass spectrometry. This pipeline builds on previous work by adding new functionality for curating and correcting annotated libraries, by validating structure annotations. Due to the high quality of our software, the reproducibility, and improved logging, we think our new pipeline has the potential to become the standard in the field for cleaning tandem mass spectrometry libraries. Graphical Abstract
Large SMILES-based Transformer Encoder-Decoder released by @IBM at #icml2024. Trained on 91 million curated SMILES from #pubchem
Interesting and entertaining keynote by Soumith Chintala at #icml2024. I agree with his positive attitude towards #opensource and #openscience.
But I don't buy the "embrace capitalism part".
Da soll noch mal jemand sagen, in Deutschland bewegt sich nichts.
New updates of my #datascience introduction course materials. Most additions and changes were made to improve and expand the #NLP chapters, going from basic string handling to TF-IDF followed by n-grams and word vectors.
Rendered version: https://florian-huber.github.io/data_science_course/
GitHub: https://github.com/florian-huber/data_science_course
Hier eine kleine Data-Science Analyse der AfD Europawahlergebnisse in Deutschland: https://medium.com/@f.huber/europawahl-2024-erkl%C3%A4rungsversuche-mittels-machine-learning-d93ec763509a
Ich versteh's noch immer nicht ... aber das ist wohl auch zuviel verlangt von dem bisschen Machine-Learning.
Notebook & Data: https://github.com/florian-huber/europawahl_2024_afd_analysis
