https://community.qlik.com/t5/Suggest-an-Idea/AutoML-Experiment-dataset-and-Deployment-update-process-life/idi-p/2048966 <P>Possibility to&nbsp;<STRONG>update the dataset</STRONG><SPAN>&nbsp;which is used in the&nbsp;</SPAN><STRONG>ML experiment&nbsp;</STRONG>would be great!</P> <P>NEXT, possibility to <STRONG>update</STRONG> (exchange) <STRONG>ML deployment</STRONG>&nbsp;by new version of the model from ML experiment would be awesome!</P> <P><SPAN>I have training data that is updated every day and (possibly) new rows are added. What I would like to achieve is to use the same ML experiment that is already setup, and just see that the training dataset was updated, run the new version and deploy a better model.</SPAN></P> <P><SPAN>But, when I look at the experiment's dataset overview, I see that it didn't update and only initial values are taken into training.</SPAN></P> <P><span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="autoML_experiment_update.PNG" style="width: 200px;"><img src="https://community.qlik.com/t5/image/serverpage/image-id/100408iD73C21645467987E/image-size/small?v=v2&amp;px=200" role="button" title="autoML_experiment_update.PNG" alt="autoML_experiment_update.PNG" /></span></P> <P>So, I have to create a new experiment for each new training dataset. This is a bit cumbersum.</P> <P><STRONG>Next level</STRONG> is to be able to also exchange the model in already created ML deployment with the new model. This will make it so much easier for app developers who use ML predictions. Currently they need to set up a new Data connection because there is always a new deployment.</P> <P>I think that both update possibilities will make it much easier to handle data science project life-cycle. Could also increase adoption of AutoML.</P> Tue, 14 Mar 2023 12:15:07 GMT RadovanOresky 2023-03-14T12:15:07Z https://community.qlik.com/t5/Suggest-an-Idea/AutoML-Experiment-dataset-and-Deployment-update-process-life/idc-p/2063638#M12718 <P>Thank you for your feedback. I can confirm that all of the items on your list are on the roadmap as part of our continued improvements to the ML Ops lifecycle. We anticipate progressively delivering these over the short and medium term.</P> Mon, 24 Apr 2023 11:59:14 GMT https://community.qlik.com/t5/Suggest-an-Idea/AutoML-Experiment-dataset-and-Deployment-update-process-life/idc-p/2063638#M12718 Steven_Pressland 2023-04-24T11:59:14Z https://community.qlik.com/t5/Suggest-an-Idea/AutoML-Experiment-dataset-and-Deployment-update-process-life/idc-p/2099943#M13217 <P>From now on, please track this idea from the Ideation portal.&nbsp;</P><P><STRONG><A title="Link to new idea" href="https://ideation.qlik.com/app/#/case/274955" target="_blank" rel="noopener">Link to new idea</A></STRONG></P><P>Meghann</P><P data-unlink="true"><EM>NOTE: Upon clicking this link 2 tabs may open - please feel free to close the one with a login page. If you <STRONG>only</STRONG> see 1 tab with the login page, please try clicking this link first: <STRONG><A title="Authenticate me!" href="#" target="_blank" rel="noopener">Authenticate me!</A></STRONG>&nbsp;t</EM><EM>hen try the link above again. Ensure pop-up blocker is off.</EM></P> Wed, 02 Aug 2023 15:06:58 GMT https://community.qlik.com/t5/Suggest-an-Idea/AutoML-Experiment-dataset-and-Deployment-update-process-life/idc-p/2099943#M13217 Meghann_MacDonald 2023-08-02T15:06:58Z https://community.qlik.com/t5/Suggest-an-Idea/AutoML-Experiment-dataset-and-Deployment-update-process-life/idc-p/2099944#M13218 Wed, 02 Aug 2023 15:06:59 GMT https://community.qlik.com/t5/Suggest-an-Idea/AutoML-Experiment-dataset-and-Deployment-update-process-life/idc-p/2099944#M13218 Ideation 2023-08-02T15:06:59Z