[Draft] Single model - Single Term changes

src/main/java/com/autotune/analyzer/kruizeObject/KruizeObject.java

@@ -17,6 +17,7 @@
 
 import com.autotune.analyzer.exceptions.InvalidValueException;
 import com.autotune.analyzer.recommendations.term.Terms;
+import com.autotune.analyzer.serviceObjects.CreateExperimentAPIObject;
 import com.autotune.analyzer.utils.AnalyzerConstants;
 import com.autotune.analyzer.utils.ExperimentTypeAware;
 import com.autotune.analyzer.utils.ExperimentTypeUtil;
@@ -26,10 +27,12 @@
 import com.autotune.utils.KruizeConstants;
 import com.autotune.utils.KruizeSupportedTypes;
 import com.autotune.utils.Utils;
+import com.fasterxml.jackson.databind.ObjectMapper;
 import com.google.gson.annotations.JsonAdapter;
 import com.google.gson.annotations.SerializedName;
 import io.fabric8.kubernetes.api.model.ObjectReference;
 
+import java.io.File;
 import java.util.HashMap;
 import java.util.List;
 import java.util.Map;
@@ -133,12 +136,16 @@ public KruizeObject() {
      @param terms A map to store the default terms with term name as the key and Terms object as the value.
      @param kruizeObject The KruizeObject for which the default terms are being set.
      */
+    // take this from json and put value here step 2
     public static void setDefaultTerms(Map<String, Terms> terms, KruizeObject kruizeObject) {
         // TODO: define term names like daily, weekly, fortnightly etc
         // TODO: add CRD for terms
+
+        // TODO : fixx this part from json objecttt figure out this
         terms.put(KruizeConstants.JSONKeys.SHORT_TERM, new Terms(KruizeConstants.JSONKeys.SHORT_TERM, KruizeConstants.RecommendationEngineConstants
                 .DurationBasedEngine.DurationAmount.SHORT_TERM_DURATION_DAYS, KruizeConstants.RecommendationEngineConstants
                 .DurationBasedEngine.DurationAmount.SHORT_TERM_DURATION_DAYS_THRESHOLD, 4, 0.25));
+        // need to discuss how will these value change based on change in term
         terms.put(KruizeConstants.JSONKeys.MEDIUM_TERM, new Terms(KruizeConstants.JSONKeys.MEDIUM_TERM, KruizeConstants
                 .RecommendationEngineConstants.DurationBasedEngine.DurationAmount.MEDIUM_TERM_DURATION_DAYS, KruizeConstants
                 .RecommendationEngineConstants.DurationBasedEngine.DurationAmount.MEDIUM_TERM_DURATION_DAYS_THRESHOLD, 7, 1));
@@ -149,6 +156,29 @@ public static void setDefaultTerms(Map<String, Terms> terms, KruizeObject kruize
         kruizeObject.setTerms(terms);
     }
 
+
+    public static void setCustomTerms(Map<String, Terms> terms, KruizeObject kruizeObject) {
+        // TODO: define term names like daily, weekly, fortnightly etc
+        // TODO: add CRD for terms
+
+        // TODO : fixx this part from json objecttt figure out this
+//        for loop for multiple terms
+
+        terms.put(kruizeObject.getRecommendation_settings().getTermSettings().getTerms().get(0), new Terms(kruizeObject.getRecommendation_settings().getTermSettings().getTerms().get(0)
+                , kruizeObject.getRecommendation_settings().getTermSettings().getTermDuration(),
+                KruizeConstants.RecommendationEngineConstants.DurationBasedEngine.DurationAmount.SHORT_TERM_DURATION_DAYS_THRESHOLD,
+                4, 0.25));
+        // need to discuss how will these value change based on change in term
+//        terms.put(KruizeConstants.JSONKeys.MEDIUM_TERM, new Terms(KruizeConstants.JSONKeys.MEDIUM_TERM, KruizeConstants
+//                .RecommendationEngineConstants.DurationBasedEngine.DurationAmount.MEDIUM_TERM_DURATION_DAYS, KruizeConstants
+//                .RecommendationEngineConstants.DurationBasedEngine.DurationAmount.MEDIUM_TERM_DURATION_DAYS_THRESHOLD, 7, 1));
+//        terms.put(KruizeConstants.JSONKeys.LONG_TERM, new Terms(KruizeConstants.JSONKeys.LONG_TERM, KruizeConstants
+//                .RecommendationEngineConstants.DurationBasedEngine.DurationAmount.LONG_TERM_DURATION_DAYS, KruizeConstants
+//                .RecommendationEngineConstants.DurationBasedEngine.DurationAmount.LONG_TERM_DURATION_DAYS_THRESHOLD, 15, 1));
+
+        kruizeObject.setTerms(terms);
+    }
+
     public String getExperimentName() {
         return experimentName;
     }

src/main/java/com/autotune/analyzer/kruizeObject/ModelSettings.java

@@ -0,0 +1,24 @@
+package com.autotune.analyzer.kruizeObject;
+
+import java.util.List;
+
+public class ModelSettings {
+
+    private List<String> models;
+
+    public List<String> getModels() {
+        return models;
+    }
+
+    public void setModels(List<String> models) {
+        this.models = models;
+    }
+
+    @Override
+    public String toString() {
+        return "ModelSettings{" +
+                "models=" + models +
+                '}';
+    }
+
+}

src/main/java/com/autotune/analyzer/kruizeObject/RecommendationSettings.java

@@ -17,6 +17,8 @@
 
 public class RecommendationSettings {
     private Double threshold;
+    private ModelSettings modelSettings;
+    private TermSettings termSettings;
 
     public Double getThreshold() {
         return threshold;
@@ -26,10 +28,28 @@ public void setThreshold(Double threshold) {
         this.threshold = threshold;
     }
 
+    public ModelSettings getModelSettings() {
+        return modelSettings;
+    }
+
+    public void setModelSettings(ModelSettings modelSettings) {
+        this.modelSettings = modelSettings;
+    }
+
+    public TermSettings getTermSettings() {
+        return termSettings;
+    }
+
+    public void setTermSettings(TermSettings termSettings) {
+        this.termSettings = termSettings;
+    }
+
     @Override
     public String toString() {
         return "RecommendationSettings{" +
                 "threshold=" + threshold +
+                ", modelSettings=" + modelSettings +
+                ", termSettings=" + termSettings +
                 '}';
     }
 }

src/main/java/com/autotune/analyzer/kruizeObject/TermSettings.java

@@ -0,0 +1,32 @@
+package com.autotune.analyzer.kruizeObject;
+
+import java.util.List;
+
+public class TermSettings {
+    private List<String> terms;
+    private String termDuration;
+
+    public List<String> getTerms() {
+        return terms;
+    }
+
+    public void setTerms(List<String> terms) {
+        this.terms = terms;
+    }
+
+    public String getTermDuration() {
+        return termDuration;
+    }
+
+    public void setTermDuration(String termDuration) {
+        this.termDuration = termDuration;
+    }
+
+    @Override
+    public String toString() {
+        return "TermSettings{" +
+                "terms=" + terms + 
+                ", singleTerm='" + termDuration + '\'' +
+                '}';
+    }
+}

src/main/java/com/autotune/analyzer/recommendations/engine/RecommendationEngine.java

@@ -67,13 +67,16 @@ public class RecommendationEngine {
     private Map<String, Terms> terms;
     private KruizeObject kruizeObject;
     private Timestamp interval_end_time;
+    private String modelName;
+//    private ModelName modelName = kruizeObject.getRecommendationSettings().getModelSettings().getModels();
 
 
     public RecommendationEngine(String experimentName, String intervalEndTimeStr, String intervalStartTimeStr) {
         this.experimentName = experimentName;
         this.intervalEndTimeStr = intervalEndTimeStr;
         this.intervalStartTimeStr = intervalStartTimeStr;
-        this.init();
+//        this.modelName = modelName;
+//        this.init(modelName);
     }
 
     private static int getNumPods(Map<Timestamp, IntervalResults> filteredResultsMap) {
@@ -123,16 +126,27 @@ private static int getNumPodsForNamespace(Map<Timestamp, IntervalResults> filter
         return (int) Math.ceil(max_pods_cpu);
     }
 
-    private void init() {
+    private void LoadRecommendationModel(String modelName) {
         // Add new models
         recommendationModels = new ArrayList<>();
         // Create Cost based model
-        CostBasedRecommendationModel costBasedRecommendationModel = new CostBasedRecommendationModel();
-        // TODO: Create profile based model
-        registerModel(costBasedRecommendationModel);
-        // Create Performance based model
-        PerformanceBasedRecommendationModel performanceBasedRecommendationModel = new PerformanceBasedRecommendationModel();
-        registerModel(performanceBasedRecommendationModel);
+        /// TODO: add if else pick models for the  get func we created -- first change
+        if("cost".equalsIgnoreCase(modelName)) {
+            CostBasedRecommendationModel costBasedRecommendationModel = new CostBasedRecommendationModel();
+            // TODO: Create profile based model
+            registerModel(costBasedRecommendationModel);
+        } else if ("performance".equalsIgnoreCase(modelName)) {
+            // Create Performance based model
+            PerformanceBasedRecommendationModel performanceBasedRecommendationModel = new PerformanceBasedRecommendationModel();
+            registerModel(performanceBasedRecommendationModel);
+        } else {
+            CostBasedRecommendationModel costBasedRecommendationModel = new CostBasedRecommendationModel();
+            // TODO: Create profile based model
+            registerModel(costBasedRecommendationModel);
+            // Create Performance based model
+            PerformanceBasedRecommendationModel performanceBasedRecommendationModel = new PerformanceBasedRecommendationModel();
+            registerModel(performanceBasedRecommendationModel);
+        }
         // TODO: Add profile based once recommendation algos are available
     }
 
@@ -168,10 +182,22 @@ public void setExperimentName(String experimentName) {
         this.experimentName = experimentName;
     }
 
+    public String getModelName() {
+        return modelName;
+    }
+
+    public void setModelName(String modelName) {
+        this.modelName = modelName;
+    }
+
     public Timestamp getInterval_end_time() {
         return interval_end_time;
     }
 
+    public String getIntervalEndTimeStr() {
+        return intervalEndTimeStr;
+    }
+
     public void setInterval_end_time(Timestamp interval_end_time) {
         this.interval_end_time = interval_end_time;
     }
@@ -291,9 +317,15 @@ public KruizeObject prepareRecommendations(int calCount, String target_cluster)
                 KruizeObject.setDefaultTerms(terms, kruizeObject);
             // set the performance profile
             setPerformanceProfile(kruizeObject.getPerformanceProfile());
+            // set custom terms
+            KruizeObject.setCustomTerms(terms, kruizeObject);
             // get the datasource
             // TODO: If no data source given use KruizeDeploymentInfo.monitoring_agent / default datasource
             String dataSource = kruizeObject.getDataSource();
+
+            setModelName(kruizeObject.getRecommendation_settings().getModelSettings().getModels().get(0));
+            LoadRecommendationModel(modelName);
+
             LOGGER.debug(String.format(KruizeConstants.APIMessages.EXPERIMENT_DATASOURCE, kruizeObject.getExperimentName(), dataSource));
 
             int maxDay = Terms.getMaxDays(terms);

src/main/java/com/autotune/analyzer/serviceObjects/Converters.java

@@ -1,14 +1,13 @@
 package com.autotune.analyzer.serviceObjects;
 
 import com.autotune.analyzer.exceptions.InvalidValueException;
-import com.autotune.analyzer.kruizeObject.ExperimentUseCaseType;
-import com.autotune.analyzer.kruizeObject.KruizeObject;
-import com.autotune.analyzer.kruizeObject.ObjectiveFunction;
-import com.autotune.analyzer.kruizeObject.SloInfo;
+import com.autotune.analyzer.kruizeObject.*;
 import com.autotune.analyzer.performanceProfiles.PerformanceProfile;
 import com.autotune.analyzer.recommendations.ContainerRecommendations;
 import com.autotune.analyzer.recommendations.NamespaceRecommendations;
+import com.autotune.analyzer.recommendations.engine.RecommendationEngine;
 import com.autotune.analyzer.recommendations.objects.MappedRecommendationForTimestamp;
+import com.autotune.analyzer.recommendations.term.Terms;
 import com.autotune.analyzer.utils.AnalyzerConstants;
 import com.autotune.common.data.ValidationOutputData;
 import com.autotune.common.data.metrics.AggregationFunctions;
@@ -70,6 +69,41 @@ public static KruizeObject convertCreateExperimentAPIObjToKruizeObject(CreateExp
                     LOGGER.debug("Experiment Type: {}", createExperimentAPIObject.getExperimentType());
                     k8sObjectList.add(k8sObject);
                 }
+                // TODO : some modification to add custom terms and models automatically here
+
+//                RecommendationSettings recommendationSettings = new RecommendationSettings();
+//                RecommendationAPIObject recommendationAPIObject = createExperimentAPIObject.getRecommendationSettings();
+//                // Process the threshold value (can be null or a default value in case it's missing)
+//                String threshold = recommendationAPIObject.getThreshold();
+//                recommendationSettings.setThreshold(threshold);
+//
+//                // Process the model settings (models is an array)
+//                ModelSettings modelSettings = recommendationAPIObject.getModelSettings();
+//                if (modelSettings != null && modelSettings.getModels() != null) {
+//                    recommendationSettings.setModels(modelSettings.getModels());
+//                }
+//
+//                // Process the term settings
+//                TermSettings termSettings = recommendationAPIObject.getTermSettings();
+//                if (termSettings != null) {
+//                    // Extract the terms array (if available)
+//                    List<String> terms = termSettings.getTerms();
+//                    recommendationSettings.setTerms(terms);
+//
+//                    // Extract individual terms with their values (like "short" => "1 day")
+//                    Map<String, String> termValues = new HashMap<>();
+//                    if (terms != null && !terms.isEmpty()) {
+//                        for (String term : terms) {
+//                            String value = termSettings.getTermValue(term); // Fetch value based on term name (e.g., "short" => "1 day")
+//                            termValues.put(term, value);
+//                        }
+//                    }
+//                    recommendationSettings.setTermValues(termValues);
+//                }
+//
+//                // Return the fully populated RecommendationSettings object
+//                return recommendationSettings;
+
                 kruizeObject.setKubernetes_objects(k8sObjectList);
                 kruizeObject.setExperimentName(createExperimentAPIObject.getExperimentName());
                 kruizeObject.setApiVersion(createExperimentAPIObject.getApiVersion());
@@ -81,7 +115,10 @@ public static KruizeObject convertCreateExperimentAPIObjToKruizeObject(CreateExp
                 kruizeObject.setExperimentType(createExperimentAPIObject.getExperimentType());
                 kruizeObject.setSloInfo(createExperimentAPIObject.getSloInfo());
                 kruizeObject.setTrial_settings(createExperimentAPIObject.getTrialSettings());
-                kruizeObject.setRecommendation_settings(createExperimentAPIObject.getRecommendationSettings());
+                TermSettings termSettings = createExperimentAPIObject.getRecommendationSettings().getTermSettings();
+                RecommendationSettings recommendationSettings = new RecommendationSettings();
+                recommendationSettings.setTermSettings(termSettings);
+                kruizeObject.setRecommendation_settings(recommendationSettings);
                 kruizeObject.setExperiment_id(createExperimentAPIObject.getExperiment_id());
                 kruizeObject.setStatus(createExperimentAPIObject.getStatus());
                 kruizeObject.setExperiment_usecase_type(new ExperimentUseCaseType(kruizeObject));