VNCoreMLFeatureValueObservation VS class VNClassificationObservation





.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty{ height:90px;width:728px;box-sizing:border-box;
}







0















I was researching on CoreML to create Machine learning app. I was reading the apple Documentation and found this two class VNCoreMLFeatureValueObservation and VNClassificationObservation. After reading documentation i was confused on what kind of model should i use these class. Also apple documentation provide different model such as mobileNet, SqueezeNet, Places205-GoogLeNet, ResNet50 and VGG16










share|improve this question





























    0















    I was researching on CoreML to create Machine learning app. I was reading the apple Documentation and found this two class VNCoreMLFeatureValueObservation and VNClassificationObservation. After reading documentation i was confused on what kind of model should i use these class. Also apple documentation provide different model such as mobileNet, SqueezeNet, Places205-GoogLeNet, ResNet50 and VGG16










    share|improve this question

























      0












      0








      0








      I was researching on CoreML to create Machine learning app. I was reading the apple Documentation and found this two class VNCoreMLFeatureValueObservation and VNClassificationObservation. After reading documentation i was confused on what kind of model should i use these class. Also apple documentation provide different model such as mobileNet, SqueezeNet, Places205-GoogLeNet, ResNet50 and VGG16










      share|improve this question














      I was researching on CoreML to create Machine learning app. I was reading the apple Documentation and found this two class VNCoreMLFeatureValueObservation and VNClassificationObservation. After reading documentation i was confused on what kind of model should i use these class. Also apple documentation provide different model such as mobileNet, SqueezeNet, Places205-GoogLeNet, ResNet50 and VGG16







      ios swift coreml






      share|improve this question













      share|improve this question











      share|improve this question




      share|improve this question










      asked Nov 24 '18 at 9:41









      SpenserSpenser

      32




      32
























          1 Answer
          1






          active

          oldest

          votes


















          0














          The VNClassificationObservation is returned when the model is a classifier. It now outputs a dictionary of [String: Double] that has the probability score for every class in the model. The models you listed are all classifiers.



          The VNCoreMLFeatureValueObservation is returned when the model is not a classifier. Instead of a dictionary, such models usually output an MLMultiArray object. You willhave to do your own post-processing to interpret the data from this kind of output.






          share|improve this answer


























          • thank you for replay. In my research on Machine Learning in iOS. i found your CoreMLHelper. i read CoreMLHelper documentation and also created prediction app with your library. I have a dought can't i use Non-maximum suppression (NMS) to create bounding box in classification observations

            – Spenser
            Nov 24 '18 at 13:56











          • NMS doesn't create bounding boxes, it only keeps the best ones. If you have a classifier output, you don't get bounding box predictions. What exactly are you trying to do?

            – Matthijs Hollemans
            Nov 25 '18 at 10:17











          • i was using ResNet50 model to classify object and i thought of adding box in that object which classifies with confidence > 0.75.

            – Spenser
            Nov 25 '18 at 12:50











          • i create object prediction app using mobile Net model, i also use you coreMLHelper. when my app start predicting objects such as person or laptop more then one bounding box appears. Is that a bug in my code or its model problem. My app does not predict accurately. i wanted my object detection to be accurate.

            – Spenser
            Nov 25 '18 at 17:04













          • Those questions are way too broad for Stack Overflow. ;-) But yes, usually multiple bounding boxes appear for each object.

            – Matthijs Hollemans
            Nov 26 '18 at 9:54












          Your Answer






          StackExchange.ifUsing("editor", function () {
          StackExchange.using("externalEditor", function () {
          StackExchange.using("snippets", function () {
          StackExchange.snippets.init();
          });
          });
          }, "code-snippets");

          StackExchange.ready(function() {
          var channelOptions = {
          tags: "".split(" "),
          id: "1"
          };
          initTagRenderer("".split(" "), "".split(" "), channelOptions);

          StackExchange.using("externalEditor", function() {
          // Have to fire editor after snippets, if snippets enabled
          if (StackExchange.settings.snippets.snippetsEnabled) {
          StackExchange.using("snippets", function() {
          createEditor();
          });
          }
          else {
          createEditor();
          }
          });

          function createEditor() {
          StackExchange.prepareEditor({
          heartbeatType: 'answer',
          autoActivateHeartbeat: false,
          convertImagesToLinks: true,
          noModals: true,
          showLowRepImageUploadWarning: true,
          reputationToPostImages: 10,
          bindNavPrevention: true,
          postfix: "",
          imageUploader: {
          brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
          contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
          allowUrls: true
          },
          onDemand: true,
          discardSelector: ".discard-answer"
          ,immediatelyShowMarkdownHelp:true
          });


          }
          });














          draft saved

          draft discarded


















          StackExchange.ready(
          function () {
          StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53456930%2fvncoremlfeaturevalueobservation-vs-class-vnclassificationobservation%23new-answer', 'question_page');
          }
          );

          Post as a guest















          Required, but never shown

























          1 Answer
          1






          active

          oldest

          votes








          1 Answer
          1






          active

          oldest

          votes









          active

          oldest

          votes






          active

          oldest

          votes









          0














          The VNClassificationObservation is returned when the model is a classifier. It now outputs a dictionary of [String: Double] that has the probability score for every class in the model. The models you listed are all classifiers.



          The VNCoreMLFeatureValueObservation is returned when the model is not a classifier. Instead of a dictionary, such models usually output an MLMultiArray object. You willhave to do your own post-processing to interpret the data from this kind of output.






          share|improve this answer


























          • thank you for replay. In my research on Machine Learning in iOS. i found your CoreMLHelper. i read CoreMLHelper documentation and also created prediction app with your library. I have a dought can't i use Non-maximum suppression (NMS) to create bounding box in classification observations

            – Spenser
            Nov 24 '18 at 13:56











          • NMS doesn't create bounding boxes, it only keeps the best ones. If you have a classifier output, you don't get bounding box predictions. What exactly are you trying to do?

            – Matthijs Hollemans
            Nov 25 '18 at 10:17











          • i was using ResNet50 model to classify object and i thought of adding box in that object which classifies with confidence > 0.75.

            – Spenser
            Nov 25 '18 at 12:50











          • i create object prediction app using mobile Net model, i also use you coreMLHelper. when my app start predicting objects such as person or laptop more then one bounding box appears. Is that a bug in my code or its model problem. My app does not predict accurately. i wanted my object detection to be accurate.

            – Spenser
            Nov 25 '18 at 17:04













          • Those questions are way too broad for Stack Overflow. ;-) But yes, usually multiple bounding boxes appear for each object.

            – Matthijs Hollemans
            Nov 26 '18 at 9:54
















          0














          The VNClassificationObservation is returned when the model is a classifier. It now outputs a dictionary of [String: Double] that has the probability score for every class in the model. The models you listed are all classifiers.



          The VNCoreMLFeatureValueObservation is returned when the model is not a classifier. Instead of a dictionary, such models usually output an MLMultiArray object. You willhave to do your own post-processing to interpret the data from this kind of output.






          share|improve this answer


























          • thank you for replay. In my research on Machine Learning in iOS. i found your CoreMLHelper. i read CoreMLHelper documentation and also created prediction app with your library. I have a dought can't i use Non-maximum suppression (NMS) to create bounding box in classification observations

            – Spenser
            Nov 24 '18 at 13:56











          • NMS doesn't create bounding boxes, it only keeps the best ones. If you have a classifier output, you don't get bounding box predictions. What exactly are you trying to do?

            – Matthijs Hollemans
            Nov 25 '18 at 10:17











          • i was using ResNet50 model to classify object and i thought of adding box in that object which classifies with confidence > 0.75.

            – Spenser
            Nov 25 '18 at 12:50











          • i create object prediction app using mobile Net model, i also use you coreMLHelper. when my app start predicting objects such as person or laptop more then one bounding box appears. Is that a bug in my code or its model problem. My app does not predict accurately. i wanted my object detection to be accurate.

            – Spenser
            Nov 25 '18 at 17:04













          • Those questions are way too broad for Stack Overflow. ;-) But yes, usually multiple bounding boxes appear for each object.

            – Matthijs Hollemans
            Nov 26 '18 at 9:54














          0












          0








          0







          The VNClassificationObservation is returned when the model is a classifier. It now outputs a dictionary of [String: Double] that has the probability score for every class in the model. The models you listed are all classifiers.



          The VNCoreMLFeatureValueObservation is returned when the model is not a classifier. Instead of a dictionary, such models usually output an MLMultiArray object. You willhave to do your own post-processing to interpret the data from this kind of output.






          share|improve this answer















          The VNClassificationObservation is returned when the model is a classifier. It now outputs a dictionary of [String: Double] that has the probability score for every class in the model. The models you listed are all classifiers.



          The VNCoreMLFeatureValueObservation is returned when the model is not a classifier. Instead of a dictionary, such models usually output an MLMultiArray object. You willhave to do your own post-processing to interpret the data from this kind of output.







          share|improve this answer














          share|improve this answer



          share|improve this answer








          edited Nov 24 '18 at 13:21

























          answered Nov 24 '18 at 11:19









          Matthijs HollemansMatthijs Hollemans

          3,2031312




          3,2031312













          • thank you for replay. In my research on Machine Learning in iOS. i found your CoreMLHelper. i read CoreMLHelper documentation and also created prediction app with your library. I have a dought can't i use Non-maximum suppression (NMS) to create bounding box in classification observations

            – Spenser
            Nov 24 '18 at 13:56











          • NMS doesn't create bounding boxes, it only keeps the best ones. If you have a classifier output, you don't get bounding box predictions. What exactly are you trying to do?

            – Matthijs Hollemans
            Nov 25 '18 at 10:17











          • i was using ResNet50 model to classify object and i thought of adding box in that object which classifies with confidence > 0.75.

            – Spenser
            Nov 25 '18 at 12:50











          • i create object prediction app using mobile Net model, i also use you coreMLHelper. when my app start predicting objects such as person or laptop more then one bounding box appears. Is that a bug in my code or its model problem. My app does not predict accurately. i wanted my object detection to be accurate.

            – Spenser
            Nov 25 '18 at 17:04













          • Those questions are way too broad for Stack Overflow. ;-) But yes, usually multiple bounding boxes appear for each object.

            – Matthijs Hollemans
            Nov 26 '18 at 9:54



















          • thank you for replay. In my research on Machine Learning in iOS. i found your CoreMLHelper. i read CoreMLHelper documentation and also created prediction app with your library. I have a dought can't i use Non-maximum suppression (NMS) to create bounding box in classification observations

            – Spenser
            Nov 24 '18 at 13:56











          • NMS doesn't create bounding boxes, it only keeps the best ones. If you have a classifier output, you don't get bounding box predictions. What exactly are you trying to do?

            – Matthijs Hollemans
            Nov 25 '18 at 10:17











          • i was using ResNet50 model to classify object and i thought of adding box in that object which classifies with confidence > 0.75.

            – Spenser
            Nov 25 '18 at 12:50











          • i create object prediction app using mobile Net model, i also use you coreMLHelper. when my app start predicting objects such as person or laptop more then one bounding box appears. Is that a bug in my code or its model problem. My app does not predict accurately. i wanted my object detection to be accurate.

            – Spenser
            Nov 25 '18 at 17:04













          • Those questions are way too broad for Stack Overflow. ;-) But yes, usually multiple bounding boxes appear for each object.

            – Matthijs Hollemans
            Nov 26 '18 at 9:54

















          thank you for replay. In my research on Machine Learning in iOS. i found your CoreMLHelper. i read CoreMLHelper documentation and also created prediction app with your library. I have a dought can't i use Non-maximum suppression (NMS) to create bounding box in classification observations

          – Spenser
          Nov 24 '18 at 13:56





          thank you for replay. In my research on Machine Learning in iOS. i found your CoreMLHelper. i read CoreMLHelper documentation and also created prediction app with your library. I have a dought can't i use Non-maximum suppression (NMS) to create bounding box in classification observations

          – Spenser
          Nov 24 '18 at 13:56













          NMS doesn't create bounding boxes, it only keeps the best ones. If you have a classifier output, you don't get bounding box predictions. What exactly are you trying to do?

          – Matthijs Hollemans
          Nov 25 '18 at 10:17





          NMS doesn't create bounding boxes, it only keeps the best ones. If you have a classifier output, you don't get bounding box predictions. What exactly are you trying to do?

          – Matthijs Hollemans
          Nov 25 '18 at 10:17













          i was using ResNet50 model to classify object and i thought of adding box in that object which classifies with confidence > 0.75.

          – Spenser
          Nov 25 '18 at 12:50





          i was using ResNet50 model to classify object and i thought of adding box in that object which classifies with confidence > 0.75.

          – Spenser
          Nov 25 '18 at 12:50













          i create object prediction app using mobile Net model, i also use you coreMLHelper. when my app start predicting objects such as person or laptop more then one bounding box appears. Is that a bug in my code or its model problem. My app does not predict accurately. i wanted my object detection to be accurate.

          – Spenser
          Nov 25 '18 at 17:04







          i create object prediction app using mobile Net model, i also use you coreMLHelper. when my app start predicting objects such as person or laptop more then one bounding box appears. Is that a bug in my code or its model problem. My app does not predict accurately. i wanted my object detection to be accurate.

          – Spenser
          Nov 25 '18 at 17:04















          Those questions are way too broad for Stack Overflow. ;-) But yes, usually multiple bounding boxes appear for each object.

          – Matthijs Hollemans
          Nov 26 '18 at 9:54





          Those questions are way too broad for Stack Overflow. ;-) But yes, usually multiple bounding boxes appear for each object.

          – Matthijs Hollemans
          Nov 26 '18 at 9:54




















          draft saved

          draft discarded




















































          Thanks for contributing an answer to Stack Overflow!


          • Please be sure to answer the question. Provide details and share your research!

          But avoid



          • Asking for help, clarification, or responding to other answers.

          • Making statements based on opinion; back them up with references or personal experience.


          To learn more, see our tips on writing great answers.




          draft saved


          draft discarded














          StackExchange.ready(
          function () {
          StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53456930%2fvncoremlfeaturevalueobservation-vs-class-vnclassificationobservation%23new-answer', 'question_page');
          }
          );

          Post as a guest















          Required, but never shown





















































          Required, but never shown














          Required, but never shown












          Required, but never shown







          Required, but never shown

































          Required, but never shown














          Required, but never shown












          Required, but never shown







          Required, but never shown







          這個網誌中的熱門文章

          Xamarin.form Move up view when keyboard appear

          Post-Redirect-Get with Spring WebFlux and Thymeleaf

          Anylogic : not able to use stopDelay()