## // Import boundaries from asset

## var aoi = ee.FeatureCollection('users/juliusadewopo/MzeTargetRegion_alt_dslv2');


## 
## function maskL8sr(image) {

##   // Bits 3 and 5 are cloud shadow and cloud, respectively.

##   var cloudShadowBitMask = (1 << 3);

##   var cloudsBitMask = (1 << 5);

##   // Get the pixel QA band.

##   var qa = image.select('pixel_qa');

##   // Both flags should be set to zero, indicating clear conditions.

##   var mask = qa.bitwiseAnd(cloudShadowBitMask).eq(0)

##                  .and(qa.bitwiseAnd(cloudsBitMask).eq(0));

##   return image.updateMask(mask);

## }

## 
## //general collection

## var l8collection = ee.ImageCollection('LANDSAT/LC08/C01/T1_SR')

##                   .filterBounds(aoi)

##                   .map(maskL8sr)

## 

## // Function to add an NDVI band, the dependent variable.

## // More details: https://landsat.usgs.gov/sites/default/files/documents/si_product_guide.pdf

## var addVI = function(image) {

##   var ndvi = image.normalizedDifference(['nir', 'red']).rename('NDVI').float();

##   var gndvi = image.normalizedDifference(['nir', 'green']).rename('GNDVI').float();

##   var grvi = image.normalizedDifference(['green', 'red']).rename('GRVI').float();

##   var gbvi = image.normalizedDifference(['blue', 'green']).rename('GBVI').float();

##     var ndmi = image.normalizedDifference(['nir', 'swir1']).rename('NDMI').float();

## var nbr2 = image.normalizedDifference(['swir1', 'swir2']).rename('NBR2').float();

##   // Compute the EVI using an expression.

##   var evi = image.expression(

##     '2.5 * ((NIR - RED) / (NIR + 6 * RED - 7.5 * BLUE + 1))', {

##       'NIR': image.select('nir'),

##       'RED': image.select('red'),

##       'BLUE': image.select('blue')

##   }).rename('EVI').float();

##    // Compute VI using expressions

## 

##   var gcvi = image.expression(

##     '(NIR/GREEN) - 1', {

##       'NIR': image.select('nir'),

##       'GREEN': image.select('green'),

##   }).rename('GCVI').float();

## 
##   // Return the masked image, with additional bands for predictors.

##   return image

##       // Add the response variables last.

##       .addBands([ndvi,gndvi, grvi,gbvi,ndmi, nbr2, evi, gcvi])

##       // Retain this metadata property for use in the chart.

##       .copyProperties(image, ['system:time_start']);

## };


## // Which bands to select:L8

## var bandNumbers = [1,2,3,4,5,6];

## var bandNames = ee.List(['blue','green','red','nir','swir1','swir2']);

## 
## // apply over the image collection

## var l8collection = l8collection.select(bandNumbers, bandNames).map(addVI);


## // Use the combined reducer - here just 25-75th

## var reducersCol = ee.Reducer.percentile({percentiles:[25]})

##                       .combine(ee.Reducer.percentile({percentiles:[50]}), null, true)

##                       .combine(ee.Reducer.percentile({percentiles:[75]}), null, true);


## var getNames = function(base, list) {

##   return ee.List(list).map(function(i) {

##     return ee.String(base).cat(i);

##   });

## };

## 
## //SEASON 1:

## var startDate = '2016-12-01';

## var endDate = '2017-3-31';

## var s1 = l8collection

##               .filterDate(startDate, endDate)

##               .reduce({

##         reducer: reducersCol

## });

## 

## var s1New = s1.select(s1.bandNames(),

##   getNames('season1_', s1.bandNames()));

## 
## //season 2:

##   var startDate = '2017-04-01';

##   var endDate = '2017-7-31';

## // Use the combined reducer - here just 25-75th

## var s2 = l8collection

##               .filterDate(startDate, endDate)

##               .reduce({

##         reducer: reducersCol

## });

## 
## var s2New = s2.select(s2.bandNames(),

##   getNames('season2_', s2.bandNames()));

## 
## //combine

## var TwoSeason = s1New.addBands(s2New);

## print(TwoSeason);


## var vizParams = {bands: ['nir', 'red', 'green'], min: 0, max: 3000};

## // to be added


## var samples = crop.merge(water).merge(soil).merge(vegetation);

## //add export


## // name of the bands

## var bands = TwoSeason.bandNames();

## 
## // extract band values from the input image

## var samplevals = TwoSeason.sampleRegions({

##   collection: samples,

##   properties: ['landcover']

##   scale: 30

## });

## 
## // define the random forest classier

## var classifier = ee.Classifier.randomForest({

##         numberOfTrees: 500,

##         variablesPerSplit: 8,

##         minLeafPopulation: 2,

##         seed: 1});

## 
## // Train the classifier

## var trained = classifier.train(samplevals, "landcover", bands);

## 
## // Classify the image with the same bands used for training.

## var classified = TwoSeason.select(bands).classify(trained);

## 
## // Display the inputs and the results.

## Map.centerObject(samples, 11);

## Map.addLayer(TwoSeason, {bands: ['B4', 'B3', 'B2'], max: 0.4}, 'image');

## Map.addLayer(classified,

##              {min: 0, max: 3, palette: ['red', 'green', 'blue']},

##              'classification');

## 
## // Export the classification result to Google drive for additional analysis

## 
## Export.image.toDrive({

##     image: classified,

##     description: "export_landsat_classification",

##     fileNamePrefix: 'landsat8_classified',

##     folder: 'GEE_export',

##     scale: 30,

##     region: aoi,

##     maxPixels: 1e13

## });


## // Step 1: Assign random numbers in preparation for a test/train split

## var seed = 100;

## samples = samples.randomColumn('random', seed);

## 
## // Join the landcover & random values with all pixels in each polygon in the training datasets.

## var samplevals = TwoSeaosn.sampleRegions({

## 	collection: samples,

## 	properties: ['landcover', 'random'],

## 	scale: 30

## });

## 
## 
## //Step 2: Perform a  30/70 test/train split using the random numbers  generated in Step 1 to assess model performance.

## 
## var training = samplevals.filterMetadata('random', 'less_than', 0.7);

## var testing = samplevals.filterMetadata('random', 'not_less_than', 0.7);

## 
## // Step 3: Train the classifier

## 
## // define the random forest classier

## var classifier = ee.Classifier.randomForest({

##         numberOfTrees: 500,

##         variablesPerSplit: 8,

##         minLeafPopulation: 2,

##         seed: 1});

## 
## // Train the classifier.

## var trained = classifier.train(training, "cropland", bands);

## 
## // Confusion matrix with independent samples

## var validation = testing.classify(trained);

## 
## // Generate accuracy metrics

## var errorMatrix = validation.errorMatrix('landcover', 'classification');

## 
## // print results

## 

## // Calculate fallowed area by pixel (0 if pixel was not fallowed)

## var areaImageSqM = ee.Image.pixelArea()

## 	.clip(roi);

## var areaImageSqKm = areaImageSqM.multiply(0.000001);

## var cropArea = cropBinary.multiply(areaImageSqKm);


## // Calculate total fallowed area in square kilometers.

## var totalCroppedArea = cropArea.reduceRegion({reducer: ee.Reducer.sum(),

## 	geometry: roi,

## 	scale: 1000});

## print('Total cropped area, sq km:', totalCroppedArea);