Error "In users/parevalo_bu/gee-ccdc-tools:ccdcUtilities/ccdc.js fit.select is not a function"

[21Sraboni]

Hi, thank you very much for your excellent work.
I'm having a problem with the new version, so I'm asking for your help. There is this issue "In users/parevalo_bu/gee-ccdc-tools:ccdcUtilities/ccdc.js
fit.select is not a function" when I want to get the result from "var ccdImage = utils.CCDC.buildCcdImage(params.Classification.changeResults, params.Classification.segs.length, params.Classification.bandNames)".
Looking forward to hearing from you back soon.
Thank you.

Regards,
Sraboni
Belgium.

[21Sraboni]

this is my code:

var utils = require('users/parevalo_bu/gee-ccdc-tools:ccdcUtilities/api');
// Define parameters
//Map.centerObject (training_data, 8);
//Map.addLayer (training_data);
// Change detection parameters
var changeDetection = {
breakpointBands: ['B3','B4','B5','B6','B7'],
tmaskBands: ['B3','B6'],
minObservations: 6,
chiSquareProbability: .99,
minNumOfYearsScaler: 1.33,
dateFormat: 2,
lambda: 20/10000,
maxIterations: 25000
};

// Classification parameters
var classification = {
bandNames: ["B1","B2","B3","B4","B5","B6","B7"],
inputFeatures: ["INTP", "SLP","PHASE","RMSE"],
coefs: ["INTP", "SLP","COS", "SIN","RMSE","COS2","SIN2","COS3","SIN3"],
ancillaryFeatures: ["ELEVATION","ASPECT","DEM_SLOPE","RAINFALL","TEMPERATURE"],
resultFormat: 'SegCollection',
classProperty: 'LC_Class',
yearProperty: 'Year',
classifier: ee.Classifier.smileRandomForest,
classifierParams: {
numberOfTrees: 150,
variablesPerSplit: null,
minLeafPopulation: 1,
bagFraction: 0.5,
maxNodes: null
},
outPath: '.',
//segs: ["S1", "S2", "S3", "S4", "S5", "S6"],
segs: ["S1", "S2", "S3", "S4", "S5", "S6"],
trainingPath: 'users/Sraboni_Sarker/sample_test1', // Modify this line with your training data path
trainingPathPredictors: 'users/Sraboni_Sarker/Predictors', // Modify this line with your training predictors path
changeResults: results,
};

var studyRegion = ee.FeatureCollection('USDOS/LSIB_SIMPLE/2017')
.filterMetadata('country_na','equals','Belgium').union()
//Map.addLayer (studyRegion, {}, 'Study Area');
//Map.centerObject (studyRegion, 8);

var params = {
start: '2018-01-01',
end: '2020-01-01',
ChangeDetection: changeDetection,
Classification: classification,
StudyRegion: studyRegion
};
/*
// Filter by date and a location in Belgium
var filteredLandsat = utils.Inputs.getLandsat()
.filterBounds(params.StudyRegion)
.filterDate(params.start, params.end)

print(filteredLandsat.size())

//Map.addLayer (filteredLandsat, {}, 'Image');
//Map.centerObject (filteredLandsat, 7);
*/
// ************************************************** //

// Load Landsat imagery
var landsatCollection = ee.ImageCollection('LANDSAT/LC08/C01/T1_SR') // Landsat 8 Surface Reflectance
.filterBounds(params.StudyRegion) // Filter by study region
.filterDate(params.start, params.end); // Filter by date range

//Map.addLayer (landsatCollection, {}, 'Image');
//Map.centerObject (landsatCollection, 7);

// Print the number of images in the Landsat collection
print('Number of Landsat images:', landsatCollection.size());

// Get one image from the Landsat collection
var sampleImage = landsatCollection.first();

// Print the bands of the sample image
print('Bands of the sample image:', sampleImage.bandNames());

// Optionally, you can sort the collection by date
var sortedLandsat = landsatCollection.sort('system:time_start');
//print('Sorted Landsat collection:', sortedLandsat);

// Use the sorted Landsat collection as the input to your change detection algorithm
params.ChangeDetection['collection'] = sortedLandsat;

//params.ChangeDetection['collection'] = filteredLandsat
params.ChangeDetection['collection'] = landsatCollection;
var results = ee.Algorithms.TemporalSegmentation.Ccdc(params.ChangeDetection);
print(results)

//*************************************//

// are they needed?? // ****************************
// Run the Temporal Segmentation algorithm

// Print the segmentation results
print('Segmentation Results:', results);

// Visualize the segmentation results on the map
//Map.addLayer(results, {}, 'Segmentation Results');

// ********************** part 2 ************************************//

/*
var params = {
Classification: {
trainingPath: 'users/Sraboni_Sarker/sample'
}
};
var trainingData = ee.FeatureCollection(params.Classification.trainingPath)
// Assuming 'params' is an object containing classification parameters
var trainingPath = params.Classification.trainingPath;

// Load the training data FeatureCollection from the specified path
var trainingData = ee.FeatureCollection(trainingPath);
*/

// // var trainingData = ee.FeatureCollection(params.Classification.trainingPath);

var trainingData = ee.FeatureCollection("users/Sraboni_Sarker/sample_test1");
// Print the size of the training data to check if it's empty
print('Number of features in training data:', trainingData.size());
//print(trainingData.first());

// // Print out more features from the training data // // Don't Delete
var numFeaturesToPrint = 480; // Change this number as needed
var moreTrainingData = trainingData.limit(numFeaturesToPrint);
print('More training data:', moreTrainingData);
print("First Feature Properties:", trainingData.first().toDictionary());

// Load training data
var trainingData = ee.FeatureCollection(params.Classification.trainingPath);
// Define the classes and their corresponding numerical values
var classMapping = {
'forest': 1,
'crop': 2,
'grass': 3,
'wetland': 4,
'settlement': 5,
'others': 6
};

// Compute histogram before remapping
var histogramBefore = ee.Dictionary(trainingData.reduceColumns({
reducer: ee.Reducer.frequencyHistogram(),
selectors: ['LC_Class']
}).get('histogram')).map(function(key, value) {
return [parseInt(key), value];
});

print('Histogram before remapping:', histogramBefore);

// Remap class labels
trainingData = trainingData.map(function(feat) {
return feat.set('LC_Class', classMapping[feat.get('LC_Class')]);
});

// Compute histogram after remapping
var histogramAfter = ee.Dictionary(trainingData.reduceColumns({
reducer: ee.Reducer.frequencyHistogram(),
selectors: ['LC_Class']
}).get('histogram')).map(function(key, value) {
return [parseInt(key), value];
});

print('Histogram after remapping:', histogramAfter);

// Set year property for all features
//trainingData = trainingData.map(function(feat) {
//return feat.set('year', 2018);
//});

// Set year property for all features
//var year = 2018; // Define the year variable
//trainingData = trainingData.map(function(feat) {
//return feat.set('year', year); // Set the 'year' property for each feature
//});

// Print the first feature to verify changes
//print('First feature after modifications:', trainingData.first());

/*
print(trainingData.aggregate_histogram('LC_Class'))
trainingData = trainingData.remap(['forest','crop','grass','wetland','settlement','others'],[1,2,3,4,5,6],'LC_Class')
print(trainingData.aggregate_histogram('LC_Class'))

trainingData = trainingData.map(function(feat) {
return feat.set('year',2014)})
/
//*****************//

// Define bands to use in classification
var bands = params.Classification.bandNames;
print (bands);

// Define coefficients to use in classification
var coefs = params.Classification.coefs;
print (coefs);

// Segment ids
var segs = params.Classification.segs;
print (segs);

// Property corresponding to year of training data
var yearProperty = params.Classification.yearProperty;

// Define path to change detection results
params.Classification.changeResults = 'users/Sraboni_Sarker/Output';

// Load ccd image stack with coefficients and change information
//var ccdImage = utils.CCDC.buildCcdImage(params.Classification.changeResults, params.Classification.segs.length, params.Classification.bandNames);
var ccdImage = utils.CCDC.buildCcdImage(params.Classification.changeResults, params.Classification.segs.length, params.Classification.bandNames);

print('CCD Image:', ccdImage);

// Finally, get ancillary topographic and climate data
var ancillary = utils.Inputs.getAncillary()

var trainingData = utils.Classification.getTrainingCoefsAtDate(
trainingData, coefs, bands, yearProperty, ancillary, ccdImage, segs);

// Filter points with no data
var testBand = params.Classification.bandNames[0] + '_' + params.Classification.coefs[0]
trainingData = trainingData.filter(ee.Filter.notNull([testBand]))

print('First training point with predictors:', trainingData.first());

Export.table.toAsset({
collection: trainingData,
description: 'trainingDataProcessed',
assetId: params.Classification.trainingPathPredictors});

//trainingData = utils.Classification.assignIds(trainingData, 'ID')

[MrShuaiB1]

var changeResults = ee.Image('users/Sraboni_Sarker/Output')