Cropland: Senegal 2022

[cnakalembe]

Start year: 2022
Start month: April

• Labeling project created
• Set 1 : Taryn, Isha, @Mirzyaaliii, Aninda, @cnakalembe, Bhanu, Kartik, Snehal, Shrey, Hitansh, Sheila, Abena
• Set 2: @mnthnx64 @Di-anaBF @hannah-rae
• Set 3: CSE - do not touch
• Data added to the repository
• Model trained
• Map made
• Expert check

[hannah-rae]

@adebowaledaniel will be exporting the data next

[hannah-rae]

✅ intercomparison results computed
next steps: 1) update CEO sample notebook to allow stratification by classification ( @MsPixels will link here the GEE script used for this before), 2) add notebook to create ensemble or other map to visualize in GEE app ( @ivanzvonkov will work on 2 #346 )

note 9/11/23: will make the land cover stratification sample separate notebook from the existing CEO sample, under the new maps folder

[MsPixels]

Senegal stratified random sampling code

[ivanzvonkov]

Here's a GEE script to take a look at the ensembled map (worldcover, glad, esri): https://code.earthengine.google.com/1a3f781cca5f0d779ff4cec47afdd035

F1 score is 0.67 +/- 0.13, which is under our 0.7 threshold, so I'll train a model to see if we can obtain a better result.

Full intercomparison report: https://github.com/nasaharvest/crop-mask/blob/master/maps/Senegal_2022/intercomparison.ipynb

[ivanzvonkov]

Trained model: #347
Results:

"test_metrics": {
    "accuracy": 0.8982,
    "f1_score": 0.4615,
    "precision_score": 0.5333,
    "recall_score": 0.4068,
},
"val_metrics": {
    "accuracy": 0.9612,
    "f1_score": 0.697,
    "precision_score": 0.7188,
    "recall_score": 0.6765,
}

Not directly comparable to intercomparison results since the evaluation set is split. The test set f1 is low and the standard deviation is probably over 0.2, so thus far ensemle-subset map is my recommendation.

[hannah-rae]

Next step: sign off from @cnakalembe if it looks good enough!

[hannah-rae]

@cnakalembe will make a qualitative check-list for CSE to evaluate the map

[hannah-rae]

re: checklist

• @ivanzvonkov make new Drive folder to house crop mask documents and add checklist
• @ivanzvonkov to review document again
• @hannah-rae to review document after above

re: map

• @ivanzvonkov updates to crop-mask workflow based on conversations with GLAD