https://matflow.celaron.com/blog/ Practical write-ups on instrument data, sparse-data modeling and formulation screening — with the limits stated. en Sun, 11 Oct 2026 00:00:00 GMT https://matflow.celaron.com/blog/arbin-res-parse-python/ https://matflow.celaron.com/blog/arbin-res-parse-python/ Sun, 11 Oct 2026 00:00:00 GMT An Arbin .res file is a Microsoft Access database in disguise. The full path from raw .res to a cycle-life plot — schema checks, channel joins, capacity recomputation — plus the three failure modes that silently corrupt the math. battery arbin instrument-data python https://matflow.celaron.com/blog/citrine-alternative-small-lab/ https://matflow.celaron.com/blog/citrine-alternative-small-lab/ Sun, 11 Oct 2026 00:00:00 GMT A practical decision guide for small R&D teams evaluating enterprise materials-informatics platforms. What we can and cannot verify about Citrine, the axes that actually matter at lab scale, and the questions to ask every platform — including Matflow. comparison gtm small-lab pricing https://matflow.celaron.com/blog/conformal-prediction-chemistry/ https://matflow.celaron.com/blog/conformal-prediction-chemistry/ Sun, 11 Oct 2026 00:00:00 GMT A cross-validated RMSE is not an error bar. Split conformal prediction turns any model into one that returns intervals with a coverage guarantee — here is the whole method in twenty lines, plus the assumption that chemistry keeps breaking. uncertainty conformal-prediction machine-learning small-data https://matflow.celaron.com/blog/ftir-baseline-correction/ https://matflow.celaron.com/blog/ftir-baseline-correction/ Sun, 11 Oct 2026 00:00:00 GMT Rolling-minimum, ALS and SNIP on real ATR spectra — how to pick the parameters, the artefacts each method creates, and the difference between a baseline correction and an ATR correction. ftir spectroscopy preprocessing python https://matflow.celaron.com/blog/gcms-peak-tables-lie/ https://matflow.celaron.com/blog/gcms-peak-tables-lie/ Sun, 11 Oct 2026 00:00:00 GMT Retention time drift moves peaks between runs, blank subtraction changes what counts as a signal, and a library match factor is a similarity score — not a probability. The preprocessing order that keeps GC-MS data honest. gcms chromatography instrument-data python https://matflow.celaron.com/blog/jcamp-dx-reader-python/ https://matflow.celaron.com/blog/jcamp-dx-reader-python/ Sun, 11 Oct 2026 00:00:00 GMT The open spectroscopy standard that every vendor exports and half the tooling misreads. What DIF, DUP and SQZ actually encode, why NTUPLES trips naive readers, and a validation checklist before you trust a parsed spectrum. spectroscopy jcamp-dx ftir nmr python https://matflow.celaron.com/blog/xrd-csv-to-phase-insight/ https://matflow.celaron.com/blog/xrd-csv-to-phase-insight/ Sun, 11 Oct 2026 00:00:00 GMT Background subtraction, Kα2 stripping, peak fitting and a Bragg-law phase check — the smallest XRD pipeline that produces defensible numbers, and the five ways it still lies to you. xrd diffraction materials python https://matflow.celaron.com/blog/biologic-mpr-python/ https://matflow.celaron.com/blog/biologic-mpr-python/ Tue, 06 Oct 2026 00:00:00 GMT Bio-Logic .mpr files drift between EC-Lab versions and hide channel metadata in binary headers. The Python route that works, the .mpt export traps, three failure modes to check for — and a no-code path to a governed dataset. battery file-formats python biologic https://matflow.celaron.com/blog/hansen-solubility-parameters/ https://matflow.celaron.com/blog/hansen-solubility-parameters/ Tue, 06 Oct 2026 00:00:00 GMT Most solubility tools give you three numbers and false confidence. A fitted HSP model with an interval and an applicability-domain flag — how to use it in five minutes, what the method can and cannot support, and when it should refuse to answer. formulations solubility screening https://matflow.celaron.com/blog/neware-ndax-cycle-life/ https://matflow.celaron.com/blog/neware-ndax-cycle-life/ Tue, 06 Oct 2026 00:00:00 GMT Neware exports hide degradation mechanisms behind cycle counts. Build the cycle table correctly, then fit competing fade laws, pick by held-out RMSE, and report remaining useful life with split-conformal intervals. battery cycle-life neware python