Export as CSV after using for loop in GEE

Question

I'm trying to use the for-loop function for a supervised classification over several dates. I have trouble exporting the accuracy metrics that are useful to me in a single CSV file. With my code I can export the data for each day separately. But I would like to have a CSV file containing all the dates for which the images have been classified. Please help me. Here is the code:

//Can be converted to import records
var Bagre = 
    /* color: #d63000 */
    /* displayProperties: [
      {
        "type": "rectangle"
      }
    ] */
    ee.FeatureCollection(
        [ee.Feature(
            ee.Geometry.Polygon(
                [[[-0.9131136144573282, 11.80001340655708],
                  [-0.9131136144573282, 11.44960967007124],
                  [-0.4901399816448282, 11.44960967007124],
                  [-0.4901399816448282, 11.80001340655708]]], null, false),
            {
              "system:index": "0"
            })]),
    other = /* color: #98ff00 */ee.FeatureCollection(
        [ee.Feature(
            ee.Geometry.Point([-0.5993166173870157, 11.584173738244099]),
            {
              "landcover": 0,
              "system:index": "0"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.7167329992229532, 11.6124239897709]),
            {
              "landcover": 0,
              "system:index": "1"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.7263460363323282, 11.676312799623682]),
            {
              "landcover": 0,
              "system:index": "2"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.6569948400432657, 11.641343901167035]),
            {
              "landcover": 0,
              "system:index": "3"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.6734743322307657, 11.580137754769975]),
            {
              "landcover": 0,
              "system:index": "4"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.7435121740276407, 11.584846396491287]),
            {
              "landcover": 0,
              "system:index": "5"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.7874574865276407, 11.531701414565942]),
            {
              "landcover": 0,
              "system:index": "6"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.8286562169963907, 11.580137754769975]),
            {
              "landcover": 0,
              "system:index": "7"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.8149233068401407, 11.635963680007837]),
            {
              "landcover": 0,
              "system:index": "8"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.7950105871135782, 11.61645950650287]),
            {
              "landcover": 0,
              "system:index": "9"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.6789674962932657, 11.487294094934224]),
            {
              "landcover": 0,
              "system:index": "10"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.6466951574260782, 11.474508871642358]),
            {
              "landcover": 0,
              "system:index": "11"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.6082430089885782, 11.54919328080972]),
            {
              "landcover": 0,
              "system:index": "12"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.5828371251995157, 11.463068970402034]),
            {
              "landcover": 0,
              "system:index": "13"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.6391420568401407, 11.603680169832181]),
            {
              "landcover": 0,
              "system:index": "14"
            })]),
    water = /* color: #0b4a8b */ee.FeatureCollection(
        [ee.Feature(
            ee.Geometry.Point([-0.5615511144573282, 11.487294094934224]),
            {
              "landcover": 1,
              "system:index": "0"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.5993166173870157, 11.499405876960067]),
            {
              "landcover": 1,
              "system:index": "1"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.6260957921917032, 11.51824539131339]),
            {
              "landcover": 1,
              "system:index": "2"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.6487550939495157, 11.533719762462855]),
            {
              "landcover": 1,
              "system:index": "3"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.6762209142620157, 11.544484039481423]),
            {
              "landcover": 1,
              "system:index": "4"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.6913271154338907, 11.537083643381099]),
            {
              "landcover": 1,
              "system:index": "5"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.7050600255901407, 11.524300681517477]),
            {
              "landcover": 1,
              "system:index": "6"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.7290926183635782, 11.549866023115023]),
            {
              "landcover": 1,
              "system:index": "7"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.7647981847698282, 11.570047541254711]),
            {
              "landcover": 1,
              "system:index": "8"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.7723512853557657, 11.586864361519334]),
            {
              "landcover": 1,
              "system:index": "9"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.6666078771526407, 11.539101952582197]),
            {
              "landcover": 1,
              "system:index": "10"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.5855837072307657, 11.491331413449927]),
            {
              "landcover": 1,
              "system:index": "11"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.5512514318401407, 11.512862801646012]),
            {
              "landcover": 1,
              "system:index": "12"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.5519380773479532, 11.482583816929104]),
            {
              "landcover": 1,
              "system:index": "13"
            }),
        ee.Feature(
            ee.Geometry.Point([-0.5759706701213907, 11.486621202893533]),
            {
              "landcover": 1,
              "system:index": "14"
            })]);

var Domain = Bagre;

var S2 = ee.ImageCollection("COPERNICUS/S2");

function maskS2clouds(image) {
  var qa = image.select('QA60');
  var cloudBitMask = 1 << 10;
  var cirrusBitMask = 1 << 11;
  var mask = qa.bitwiseAnd(cloudBitMask).eq(0)
      .and(qa.bitwiseAnd(cirrusBitMask).eq(0));
  return image.updateMask(mask).divide(10000);
}

var gcp = other.merge(water);
var gcp_random = gcp.randomColumn();
var trainingGcp = gcp_random.filter(ee.Filter.lt('random', 0.7));
var validationGcp = gcp_random.filter(ee.Filter.gte('random', 0.7));

// List the years, months, and dates
var years = [2021];
var months = [1,1,1,1,1,1];
var days1 = [1,6,11,16,21,26];
var i;
var j;
for (i = 0; i<years.length;i++){
  for(j = 0; j<months.length;j++){
    var dayOfInterest = ee.Date(years[i] + "-" + months[j]  + "-" + days1[j]);
    var loop = S2
    .filter(ee.Filter.date(dayOfInterest, dayOfInterest.advance(1, 'day')))
    .filter(ee.Filter.bounds(Domain))
    .map(maskS2clouds)
    .select('B.*')
    .median()
    .clip(Domain);

// Extract Digital Numbers of pixels in training polygons
    var training = loop.sampleRegions({
      collection: trainingGcp,
      properties: ['landcover'],
      scale: 10,
      tileScale: 16,
    });
    
// Train a classifier.
    var classifier = ee.Classifier.smileRandomForest(100)
    .train({
      features: training,  
      classProperty: 'landcover',
      inputProperties: loop.bandNames()
    });
    
// Classify the satellite image
        var classified = loop.classify(classifier);

/* Accuracy Assessment*/
    var test = classified.sampleRegions({
    collection: validationGcp,
    properties: ['landcover'],
    scale: 10,
    tileScale: 16,
    });
    var testConfusionMatrix = test.errorMatrix('landcover', 'classification');

// Export accuracy assessment metrics as CSV
    var fc = ee.FeatureCollection([
    ee.Feature(null, {
    'Producers Accuracy': testConfusionMatrix.producersAccuracy().get([1,-1]).multiply(100),
    'Consumers Accuracy': testConfusionMatrix.consumersAccuracy().get([0,1]).multiply(100),
    'Overall Accuracy': testConfusionMatrix.accuracy().multiply(100),
    'Kappa coefficient': testConfusionMatrix.kappa().multiply(100),
    'Days': dayOfInterest.format('YYYY-MM-dd')
    })
    ]);
    Export.table.toDrive({
    collection: fc,
    description: 'Bagre_Area_Sq_Km' + "_" + years[i] + "-" + months[j]  + "-" + days1[j],
    folder: 'earthengine',
    fileFormat: 'CSV',
    selectors:['Days','Overall Accuracy', 'Producers Accuracy', 'Consumers Accuracy', 'Kappa coefficient']
    });
  }
}

Answer 1

You can do that by saving all the features inside an ee.List and then casting that list into a ee.FeatureCollection. For that, you need to create an empty list before the for loops and create an auxiliary variable aux. Then, instead of exporting each ee.Feature as an ee.FeatureCollection, save it inside the list. Finally, after the loops have ended you should cast the list into a ee.FeatureCollection and export it. As a small piece of advice, I strongly suggest you use map instead of for to perform the loops, since map is the recommended function to make loops in the server-side. Using the map function can also simplify some parts of the code, since mapping over an ee.List will automatically organize the results into an ee.List.

// Part before the for loops
// Start empty list
var expList = ee.List([]);
// Start aux counter
var aux = 0;

// Final part of the for loops
// Export accuracy assessment metrics as CSV
    var fc = 
    ee.Feature(null, {
    'Producers Accuracy': testConfusionMatrix.producersAccuracy().get([1,-1]).multiply(100),
    'Consumers Accuracy': testConfusionMatrix.consumersAccuracy().get([0,1]).multiply(100),
    'Overall Accuracy': testConfusionMatrix.accuracy().multiply(100),
    'Kappa coefficient': testConfusionMatrix.kappa().multiply(100),
    'Days': dayOfInterest.format('YYYY-MM-dd')
    });
expList = expList.insert(aux,fc);
aux = aux+1;

// After the for loops
// Create FeatureCollection from list with features
var exportFC = ee.FeatureCollection(expList);

// Take a look at the console
print('exportFC', exportFC);

Export.table.toDrive({
    collection: exportFC,
    description: 'Bagre_Area_Sq_Km' + "_" + "all",
    folder: 'earthengine',
    fileFormat: 'CSV',
    selectors:['Days','Overall Accuracy', 'Producers Accuracy', 'Consumers Accuracy', 'Kappa coefficient']
    });