Model Migration: from Excel to R
summary
An attempted migration that copies Excel logic into a single R script consumes all memory and crashes the server after 10–15 million iterations of a simulation that produces 30 risk classes per run. The fix is to stop replicating spreadsheet design, define the model as A→B, encapsulate operations as functions, preallocate list slots for simulations, avoid needless copying inside loops, and postpone parallelism—start with small samples, profile, then scale.
AI-written summary · about
chapters (4)
related talks
15mAdvocating for Automation: Adapting Current Tools in Environmental Science with R
35mBetter Research Planning Through Simulation
25mSql Server and R for Real-time Predictions
57mGood Practices for Applied Machine Learning
22m. Cultivating Your Own R Ecosystem as a Solo Contributor
54mData Analysis with R and Julia
46mElegant R: Tips from a Software Developer
43mInteroperability between R Language & other Languages