Justin Gosses

Currently, things like fining all available curves in wells and finding neighboring wells are done with just the input files needed for those tasks, then the results merged. If someone...

Expand number of places Dask is used to parallerize functions. Generally try to speed things up. Dask was originally included than taken out as debugging was taking too long. Could...

debug why some wells can't be loaded and are ignored

problem in features_runner.py len(df_test5) 1302634 /Users/justingosses/anaconda/envs/MannvilleDask2/lib/python3.6/site-packages/pandas/core/generic.py:1996: PerformanceWarning: your performance may suffer as PyTables will pickle object types that it cannot map directly to c-types [inferred_type->mixed,key->block3_values] [items->['UWI', 'trainOrTest', 'Neighbors_Obj', 'class_DistFrPick_TopTarget', 'class_DistFrPick_TopHelper',...

https://github.com/JustinGOSSES/MannvilleGroup_Strat_Hackathon/issues

Add example of mapping various attributes to plot.py - Error - Range of depth of top X number of depth predictions by probability - Actual top depth - Unit thickness...

Run with a different top than Top McMurry and post results here then suggest leader table for results in readme.

Add catBoost and lightGBM to XGBoost functionality to increase speed of model training.