Browse Source

add few text

master
Olivier Courtin 4 months ago
parent
commit
6448f1e3fb
1 changed files with 20 additions and 2 deletions
  1. +20
    -2
      docs/qa.md

+ 20
- 2
docs/qa.md View File

@@ -1,5 +1,8 @@
# Neat-EO Quality Analysis

This tutorial allow you in few hours (depending on your hardware),
to see how you can validate Quality of a Dataset, easily and at scale.


Check Neat-EO.pink installation and GPU
---------------------------------------
@@ -84,13 +87,28 @@ neo compare --cover cover.csv --mode stack --images gl/images gl/labels gl/masks
neo compare --cover cover.csv --mode list --labels gl/labels --masks gl/masks --max Building QoD 0.80 --out gl/outliers.cover.csv
neo cover --cover gl/outliers.cover.csv --type geojson --out gl/compare/tiles.json
```

- Pink pixels: predicted by model
- Green pixels: labels
- Grey pixels: both model and labels agree
- Pink squares: significant differences between both

<a href="https://datapink.net/neo/qa/compare/"><img src="https://datapink.io/datapink/neat-EO/raw/branch/master/docs/img/qa/compare.png" /></a>

Filter training DataSet by selecting then removing unwanted outliers
--------------------------------------------------------------------
Manually Filter training DataSet by selecting unwanted outliers
---------------------------------------------------------------
```bash
neo compare --cover gl/outliers.cover.csv --mode side --images gl/images gl/compare --out gl/compare_outliers
```

It would take about an hour to a human, to do so.
Here a possible result:
```bash
wget -O gl/remove.cover.csv https://datapink.net/neo/qa/remove.cover.csv
```

Then filter our dataset ro remove the unwanted outliers.
```bash
neo subset --dir gl/images --cover gl/cover.csv --out gl/filter/images
neo subset --dir gl/labels --cover gl/cover.csv --out gl/filter/labels
neo subset --dir gl/filter/images --delete --cover gl/remove.cover.csv


Loading…
Cancel
Save