1. Investigate sim-duration, vm depth, vm area correlations with CD and Mw for NZ-wide validation
2. Enabling multiple empirical IM predictions
3. Run simulations for validation dataset (400m → 200m/100m)

1. Jason

Robin

1. Done
2. Unblocked
3. Workflow verification underway

1) https://uceqeng.slack.com/archives/C0GBS5GQ3/p1584484502019100 - updated plots with 40 km minimum rrup

1. pre-production testing
2. Fix CUDA HF bug (file io suspected)
3. Non uniform grid generation (wrap up quickly)
4. Slack alert (to mitigate the HF hiccup issue)
   
   

Claudio/Jason/Sung

3. Production Grid running -

4. Done

Cybershake v20p4

TeamGantt

1. Pre Production Testing

3) Non Uniform Grid 20.3

4) Slack API sends msg to #cybershake_runs (private)

1. GeoNET code Python3-ize (LP)
2. Automated workflow for empirical
3. gmsim-template params test.

1) Jason (background)

2) Sung

3) Jonney

1. GM Selection for Empirical
2. Automate documentation

3. 3 points from BB's comment – NZGS v2 + DS disagg

Claudio, Jason, Daniel

2) Background task (Jason)

3) Done

Roadmap (scientific functionality list)

Production - TODO (longer term tasks)

Benchmark (Compare with output from R code)

1. 37 stations
2. non-uniform grid
3. Time estimate for entire country

Viktor

1. custom location function, runs mvn for Ahdi and Yong followed by weighted join. Results in link (attached spreadsheet).
2. .
3. Runs overnight ~8-10hours

1. Testing Cythonized Elastic/Inelastic IM (low priority)
   
   

1) James

1) NN - GMM

• Implement an initial basic pipeline with some NN config + flexible feature selection & preprocessing

2) GM Classifier

a) Mike to classify quality, minimum frequency

b) Compare against original on full val dataset

c) Grid search

d) Active learning (subduction & fmin?)

Claudio

1) Some discussions, otherwise no changes

2)
a) Mike has classified initial 200 records, selected via M-Rrup plot
b) Done apart from Fisher-Z transformation (low priority)
c) No progress

d) Exploratory work

1. Filter median empirical aggregations (+manual selection)
   

Jason

James

1. SimAtlas simulation+animation: 
   : Test auto workflow with batch 4. (total 100 faults)
   
   
2. Investigate AWS/Azure container solutions
   
   
   
   

Sung

Jonney

Sung

1.SimAtlas simulation+animation

2.