epimodel / epiabc Goto Github PK
View Code? Open in Web Editor NEWApproximate Bayesian Computation for EpiModel on High-Performance Computing Clusters
Approximate Bayesian Computation for EpiModel on High-Performance Computing Clusters
Please move these over from your fx.R
file to here, and the update the calibration workflow to use the package.
should leave the .out file blank if sucess, currently has two nulls if done
if no files found then specify
When running EpiABC to estimate EpiModel parameters, EpiABC produce sample posteriors whose mean accurately hits target statistics. However, when using estimated model parameters, EpiModel has not been able to replicate this performance in practice, e.g. fails to achieve equilibrium within a prespecified time frame.
I used EpiABC to calibrate my model to the prevalence of diagnosed infection (scripts here), but when I simulated using the means of the posterior distribution, I found that true prevalence (i.prev, in blue) instead of diagnosed prevalence (in red) is matching the target value. I calibrated to prevalence in years 70-79 (prevalence is still fluctuating at this point, which is related to a separate issue that I'm addressing now, but the mean true prevalence in these years is pretty closet to the target value).
I couldn't find anything in my scripts that would explain this, so on a whim, I tried defining i.prev
in my model to be the same as diagnosed prevalence, and I defined a new variable for true prevalence. With this change, diagnosed prevalence (in red) is close(ish) my calibration target value of 0.0819 in years 70-79.
This suggests to me that there may be something hard-coded in the EpiABC functions to fit to "i.prev" by default.
Like this
sbatch -p ckpt -A csde-ckpt --array=1-72 --job-name=wave0 --ntasks-per-node=28 --mem=100G --time=1:00:00 --export=ALL,wave=0 runsim.sh
sbatch -p ckpt -A csde-ckpt --array=1-58 --depend=afterany:582211 --job-name=wave1 --ntasks-per-node=28 --mem=100G --time=1:00:00 --export=ALL,wave=1 runsim.sh
sbatch -p ckpt -A csde-ckpt --array=1-58 --depend=afterany:582283 --job-name=wave2 --ntasks-per-node=28 --mem=100G --time=1:00:00 --export=ALL,wave=2 runsim.sh
sbatch -p ckpt -A csde-ckpt --array=1-58 --depend=afterany:582295 --job-name=wave3 --ntasks-per-node=28 --mem=100G --time=1:00:00 --export=ALL,wave=3 runsim.sh
sbatch -p ckpt -A csde-ckpt --array=1-58 --depend=afterany:582306 --job-name=wave4 --ntasks-per-node=28 --mem=100G --time=1:00:00 --export=ALL,wave=4 runsim.sh
sbatch -p ckpt -A csde-ckpt --array=1-58 --depend=afterany:582345 --job-name=wave5 --ntasks-per-node=28 --mem=100G --time=1:00:00 --export=ALL,wave=5 runsim.sh
Can do lookup of depend based on jobname with this apparently:
sbatch --dependency=$(squeue --noheader --format %i --name <JOB_NAME>)
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.