ANL-CEEESA/RELOG
1
0
Fork 0
mirror of https://github.com/ANL-CEEESA/RELOG.git synced 2025-12-06 07:48:50 -06:00
Code Issues Packages Projects Releases Wiki Activity

Dist: Drop NaN in training dataset

Browse Source
This commit is contained in:
Alinson S. Xavier
2023-01-24 10:27:41 -06:00
parent d1f6796c96
commit 79748e3c13
1 changed files with 5 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/graph/dist.jl
View File

@@ -40,7 +40,8 @@ function _calculate_distance(
end
# Fit kNN model
df = DataFrame(CSV.File(csv_filename))
df = DataFrame(CSV.File(csv_filename, missingstring="NaN"))
dropmissing!(df)
coords = Matrix(df[!, [:source_lat, :source_lon, :dest_lat, :dest_lon]])'
metric.ratios = Matrix(df[!, [:ratio]])
metric.tree = KDTree(coords)
@@ -53,5 +54,7 @@ function _calculate_distance(
# Predict ratio
idxs, _ = knn(metric.tree, [source_lat, source_lon, dest_lat, dest_lon], 5)
ratio_pred = mean(metric.ratios[idxs])
return round(dist_euclidean * ratio_pred, digits = 3)
dist_pred = round(dist_euclidean * ratio_pred, digits = 3)
isfinite(dist_pred) || error("non-finite distance detected: $dist_pred")
return dist_pred
end