Michał M. Więcław

[LeanCode 💙 @LeanCodePl]

Flutter CTO Survey 2024 💙

Calling all CTOs, CIOs, and Senior Tech Executives who have integrated Flutter into your organizations!

We need your feedback to help shape the future of Flutter.

SURVEY LINK: https://t.co/T0r7EWmZRB

[LeanCode 💙 @LeanCodePl]

🌟True story: what it takes to deliver newsfeed feature. Spoiler alert: it wasn’t easy. In our recent project, we enhanced a #Flutter mobile app with dynamic social media elements. We chose @getstream_io as our ally. How did it go? See our 🧵. Full story: https://t.co/ZAhUIlkrKi

[JOSS @JOSS_TheOJ]

Just published in JOSS: '' https://t.co/jlvLH3cKOv

[Michał M. Więcław @mmwieclaw]

Extremely happy to see my software published in @JOSS_TheOJ! And seeing that it misses a title in the tweet is really something else. 🤣

[JOSS @JOSS_TheOJ]

Just published in JOSS: '' https://t.co/jlvLH3cKOv

[Michał M. Więcław @mmwieclaw]

@ghutchis @greensteinbri @IOC_PAS @FurChem I just thought about another thing that might be adding up to the difference: data access model. Tesliper is using dicts for storage, while cclib uses object's attributes, which is a bit slower. (About 30%, I've seen somewhere.)

[Michał M. Więcław @mmwieclaw]

I'm presenting the software I've created to facilitate the simulation of optical spectra of organic compounds.
Happy to answer any questions!

Tesliper: Spectral Simulations Simplified
#RSCPoster #RSCDigital @IOC_PAS @FurChem
#RSCPosterPitch & PDF in the comments!

[Michał M. Więcław @mmwieclaw]

@pk_kraus @ghutchis @greensteinbri @IOC_PAS @FurChem @MolSsi Not sure if I will adopt QCSchema to tesliper, but maybe I'll ask anyway.

[Michał M. Więcław @mmwieclaw]

@Silvannasa On my defense 😉, QC and simulations are not my main domains, I'm an organic chemist by training. Can't figure out everything, apparently. 🤷

[Michał M. Więcław @mmwieclaw]

@Silvannasa At first I wanted to include NMR simulations as well, but honestly, it was over my head. 😅 Probably will stay with optical spectra for now.

[Michał M. Więcław @mmwieclaw]

@Silvannasa Note however, that tesliper was not created in parallelism in mind, so you won't get any performance boost.

[Michał M. Więcław @mmwieclaw]

@Silvannasa Thanks! If the cluster has Python 3.6+ installed and allows you to `pip install tesliper` then I see no problem. I've included a scripting tutorial in the documentation if you'd like to give it a try. :)

[Michał M. Więcław @mmwieclaw]

@ghutchis @greensteinbri @IOC_PAS @FurChem I will look into that, thanks for the suggestion!

[Michał M. Więcław @mmwieclaw]

@LetciaMariaPeq1 @IOC_PAS @FurChem Thank you! :)

[Michał M. Więcław @mmwieclaw]

@pk_kraus @ghutchis @greensteinbri @IOC_PAS @FurChem @MolSsi I didn't know about QCSchema project, thank you for the suggestion! I've quickly skimmed through the specification and didn't see spectral data, did I missed it?

[Michał M. Więcław @mmwieclaw]

@ghutchis @greensteinbri @IOC_PAS @FurChem Also, thank you for stopping by and taking a look at my project, I really appreciate it! 😁

[Michał M. Więcław @mmwieclaw]

@ghutchis @greensteinbri @IOC_PAS @FurChem I didn't investigate on this matter much, but my guess is: 1) cclib's overhead of finding appropriate parser; 2) my parser is more specialized and ignores many things.

[Michał M. Więcław @mmwieclaw]

@ghutchis @greensteinbri @IOC_PAS @FurChem I know about cclib, it's a great package! I was thinking about providing a bridge to allow using it as an alternative parser.

[Michał M. Więcław @mmwieclaw]

@anatmilo @IOC_PAS @FurChem However, I'm thinking about adding support for reading .sdf and/or .mol files. If you'd like to see this feature in Tesliper, let me know!

[Michał M. Więcław @mmwieclaw]

@anatmilo @IOC_PAS @FurChem Thanks! Yes, conformational search needs to be done with other means. Tesliper is mostly designed for batch processing of Gaussian output files (.log/.out).

[Michał M. Więcław @mmwieclaw]

@greensteinbri @IOC_PAS @FurChem Unfortunately, only Gaussian for now. I'm willing to add support for other QC packages, however I have experience only with Gaussian. I'm open to collaborate!

[Michał M. Więcław @mmwieclaw]

@alanqf22 @theochemmerida @CINVESTAVMERIDA @VioletaDzib @CarlosLoast @BacktotheFuture Looks great and the presentation is 🔥! I'm only wondering why did you decide to keep the source code closed?