# Senior Data Science: Safe Aging with SPHERE **URL:** https://community.drivendata.org/c/senior-data-science-sphere/16.md [Latest](https://community.drivendata.org/latest.md) · [Categories](https://community.drivendata.org/categories.md) --- ## [About the Senior Data Science: Safe Aging with SPHERE category](https://community.drivendata.org/t/about-the-senior-data-science-safe-aging-with-sphere-category/632)
**Author:** [@isms](https://community.drivendata.org/u/isms)\ **Replies:** 0
This category is for posts about the Senior Data Science: Safe Aging with SPHERE competition. Please keep all posts here specific to the competition. --- ## [What approaches did people use?](https://community.drivendata.org/t/what-approaches-did-people-use/736)
**Author:** [@mlearn](https://community.drivendata.org/u/mlearn)\ **Replies:** 11\ **Last updated:** [August 14, 2016, 11:11am UTC](https://community.drivendata.org/t/what-approaches-did-people-use/736 "2016-08-14T11:11:33Z")
This was a fun competition but unfortunately I didn't have enough time to fully engage with it. What worked? I'd have loved to have had time to try a recurrent neural network on this data but instead I only managed … --- ## [\#6 solution blog post](https://community.drivendata.org/t/6-solution-blog-post/743)
**Author:** [@aakansh9](https://community.drivendata.org/u/aakansh9)\ **Replies:** 1\ **Last updated:** [August 11, 2016, 8:06am UTC](https://community.drivendata.org/t/6-solution-blog-post/743 "2016-08-11T08:06:57Z")
I wrote down a blog post with brief description of approach towards the end. It also has a link to a gist with code on how I customized xgboost for this data. --- ## [Any xgbooster here?](https://community.drivendata.org/t/any-xgbooster-here/729)
**Author:** [@sonam](https://community.drivendata.org/u/sonam)\ **Replies:** 9\ **Last updated:** [July 27, 2016, 9:04am UTC](https://community.drivendata.org/t/any-xgbooster-here/729 "2016-07-27T09:04:10Z")
anyone using xgboost or wanna team up ? --- ## [Future Data Usage](https://community.drivendata.org/t/future-data-usage/725)
**Author:** [@aakansh9](https://community.drivendata.org/u/aakansh9)\ **Replies:** 1\ **Last updated:** [July 24, 2016, 6:33pm UTC](https://community.drivendata.org/t/future-data-usage/725 "2016-07-24T18:33:56Z")
To predict activity for a time window ( t-1, t ) can we use features created from data collected after time t ? Additionally, is it allowed to use the prior knowledge from the script provided to participants ? --- ## [Meaning of the Targets are Unclear](https://community.drivendata.org/t/meaning-of-the-targets-are-unclear/720)
**Author:** [@fhidalgo](https://community.drivendata.org/u/fhidalgo)\ **Replies:** 1\ **Last updated:** [July 21, 2016, 8:40am UTC](https://community.drivendata.org/t/meaning-of-the-targets-are-unclear/720 "2016-07-21T08:40:37Z")
Hey, Can someone clarify what the targets are? The text states, "The mean duration of each label within this window is used as the target variable." Does this mean that if it's .333 for walk, .333 for stand, and .33… --- ## [Questions about Data and Study Design](https://community.drivendata.org/t/questions-about-data-and-study-design/663)
**Author:** [@rmrost01](https://community.drivendata.org/u/rmrost01)\ **Replies:** 6\ **Last updated:** [July 21, 2016, 8:51am UTC](https://community.drivendata.org/t/questions-about-data-and-study-design/663 "2016-07-21T08:51:58Z")
The paper describes scripted data and non-scripted data along with a link to the script but the page can not be found! Which files are scripted and which files aren't scripted? What is the script? How many participants p… --- ## [Question about annotators in test set](https://community.drivendata.org/t/question-about-annotators-in-test-set/713)
**Author:** [@lihang00](https://community.drivendata.org/u/lihang00)\ **Replies:** 1\ **Last updated:** [July 17, 2016, 9:59am UTC](https://community.drivendata.org/t/question-about-annotators-in-test-set/713 "2016-07-17T09:59:56Z")
Quick question, are we allowed to use annotators of series (who annotate which series) in data set? @isms @ntwomey are you organizer of this competition? Do you know who can confirm my question? --- ## [External data usage](https://community.drivendata.org/t/external-data-usage/697)
**Author:** [@dremovd](https://community.drivendata.org/u/dremovd)\ **Replies:** 1\ **Last updated:** [July 11, 2016, 9:43am UTC](https://community.drivendata.org/t/external-data-usage/697 "2016-07-11T09:43:42Z")
Hello! It is OK to use publicly available external data by rules (as far as I understood). Do you think it is possible for such kind of contest? Did you find anything useful? --- ## [\[urgent\] Test files with no acceleration inputs](https://community.drivendata.org/t/urgent-test-files-with-no-acceleration-inputs/691)
**Author:** [@marcvshdr](https://community.drivendata.org/u/marcvshdr)\ **Replies:** 3\ **Last updated:** [July 8, 2016, 9:52am UTC](https://community.drivendata.org/t/urgent-test-files-with-no-acceleration-inputs/691 "2016-07-08T09:52:24Z")
Hello, it seems that there are multiple files in the test data directory with no acceleration inputs. For example, public\_data/test/00015/acceleration.csv has no rows, even though in the sample submission files it is l… --- ## [Is your performance on training data quite different from that on testing data?](https://community.drivendata.org/t/is-your-performance-on-training-data-quite-different-from-that-on-testing-data/653)
**Author:** [@liuxixi](https://community.drivendata.org/u/liuxixi)\ **Replies:** 5\ **Last updated:** [July 8, 2016, 9:04am UTC](https://community.drivendata.org/t/is-your-performance-on-training-data-quite-different-from-that-on-testing-data/653 "2016-07-08T09:04:28Z")
I found that my performance on testing data given by the leaderboard is much lower than performance on training data. Is it because that the testing data is much more challenging than the training data? I just want to ma… --- ## [Weighted Brier score](https://community.drivendata.org/t/weighted-brier-score/641)
**Author:** [@rpmcruz](https://community.drivendata.org/u/rpmcruz)\ **Replies:** 4\ **Last updated:** [June 9, 2016, 1:31pm UTC](https://community.drivendata.org/t/weighted-brier-score/641 "2016-06-09T13:31:56Z")
The competitions uses some form of multi-class weighted Brier score. My implementation seems to be a bit off (or it could be my cross-validation which is wack). Can someone see if there is something wrong: # get w\_c … --- ## [Use this project template for your analysis!](https://community.drivendata.org/t/use-this-project-template-for-your-analysis/634)
**Author:** [@bull](https://community.drivendata.org/u/bull)\ **Replies:** 0\ **Last updated:** [June 6, 2016, 10:23pm UTC](https://community.drivendata.org/t/use-this-project-template-for-your-analysis/634 "2016-06-06T22:23:37Z")
Hey y'all, DrivenData recently opensourced our data science project template: Cookiecutter Data Science You'll get mega street cred and internet points and rainbows and kittens if you use this project structure for yo…