# Is it possible to reinitialize the model at a certain date? **URL:** \ **Category:** Elasticsearch\ **Tags:** elastic-stack-machine-learning\ **Created:** [January 2, 2019, 9:57am UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627 "2019-01-02T09:57:51Z")\ **Posts on this page:** 11\ **Page:** 1
**Author:** ![dao](https://avatars.discourse-cdn.com/v4/letter/d/a6a055/32.png) [@dao](https://discuss.elastic.co/u/dao)\ **Post date:** [January 2, 2019, 9:57am UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627/1 "2019-01-02T09:57:51Z")
Hello, I am monitoring sensor data with ML. Each time the sensor is replaced, the values jumps, that's something I am OK with. But after the jump, the model bounds are getting wider. I don't want this, and I want to restart from scratch. ![image](https://us1.discourse-cdn.com/elastic/original/3X/f/8/f8cc1dc59af43a5c58e95a661d43c3b8488d118a.png) My job is a partition\_field\_name one, means that the job is split by sensor. So I want to reset only 1 model, not the others thank's for your help ---
**Author:** ![richcollier](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/richcollier/32/115035_2.png) [@richcollier](https://discuss.elastic.co/u/richcollier)\ **Post date:** [January 2, 2019, 2:28pm UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627/2 "2019-01-02T14:28:24Z")
You cannot reset the model for an individual partition - only the job as a whole Just so you know - the model should relatively quickly adapt to the new sensor's behavior. ML is designed for this so that there's less burden on model maintenance. By the way, do you really only get sensor information once per day as your screenshot suggests? ---
**Author:** ![dao](https://avatars.discourse-cdn.com/v4/letter/d/a6a055/32.png) [@dao](https://discuss.elastic.co/u/dao)\ **Post date:** [January 2, 2019, 2:43pm UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627/3 "2019-01-02T14:43:33Z")
Around 1 per day, yes. Is it a problem? ---
**Author:** ![richcollier](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/richcollier/32/115035_2.png) [@richcollier](https://discuss.elastic.co/u/richcollier)\ **Post date:** [January 2, 2019, 3:04pm UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627/4 "2019-01-02T15:04:28Z")
Nope, not a problem. Was just curious as to how "real-time" your data actually was. ---
**Author:** ![dao](https://avatars.discourse-cdn.com/v4/letter/d/a6a055/32.png) [@dao](https://discuss.elastic.co/u/dao)\ **Post date:** [January 2, 2019, 3:37pm UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627/5 "2019-01-02T15:37:41Z")
Anyway, I do not find that the model adapts very quickly. Here is an other example. As you can see it took like for the model to 15 days to drop down and the accuracy is very low for seveal months from this moment. ![image.png](https://us1.discourse-cdn.com/elastic/original/3X/3/7/378507d79a0f2c1ecb402373641b7f2e3e817945.png) ---
**Author:** ![richcollier](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/richcollier/32/115035_2.png) [@richcollier](https://discuss.elastic.co/u/richcollier)\ **Post date:** [January 2, 2019, 3:41pm UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627/6 "2019-01-02T15:41:10Z")
What version are you using? ---
**Author:** ![richcollier](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/richcollier/32/115035_2.png) [@richcollier](https://discuss.elastic.co/u/richcollier)\ **Post date:** [January 4, 2019, 3:57pm UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627/7 "2019-01-04T15:57:18Z")
@dao - What version are you using? ---
**Author:** ![dao](https://avatars.discourse-cdn.com/v4/letter/d/a6a055/32.png) [@dao](https://discuss.elastic.co/u/dao)\ **Post date:** [January 4, 2019, 4:14pm UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627/8 "2019-01-04T16:14:10Z")
Version: 6.5.1 regards, ---
**Author:** ![dao](https://avatars.discourse-cdn.com/v4/letter/d/a6a055/32.png) [@dao](https://discuss.elastic.co/u/dao)\ **Post date:** [January 4, 2019, 4:35pm UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627/9 "2019-01-04T16:35:23Z")
Sorry, I have 2 clusters, one in 6.2.2 and the other in 6.5.1 ---
**Author:** ![richcollier](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/richcollier/32/115035_2.png) [@richcollier](https://discuss.elastic.co/u/richcollier)\ **Post date:** [January 4, 2019, 5:55pm UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627/10 "2019-01-04T17:55:52Z")
Ok thanks - i ask because in v6.4 - a specific feature was put in to allow the model to adjust more quickly after a state change: [https://www.elastic.co/blog/designing-for-change-in-elastic-machine-learning](https://www.elastic.co/blog/designing-for-change-in-elastic-machine-learning) However, even so, with only daily samples, adjusting to change still may seem like a long time (in terms of time), but in terms of the number of samples, it's actually not that much. ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [February 1, 2019, 5:55pm UTC](https://discuss.elastic.co/t/is-it-possible-to-reinitialize-the-model-at-a-certain-date/162627/11 "2019-02-01T17:55:53Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.