# Calculating the score **URL:** \ **Category:** Warm Up: Predict Blood Donations\ **Created:** [December 30, 2018, 12:47pm UTC](https://community.drivendata.org/t/calculating-the-score/2965 "2018-12-30T12:47:29Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![Thakshi](https://avatars.discourse-cdn.com/v4/letter/t/f9ae1b/32.png) [@Thakshi](https://community.drivendata.org/u/Thakshi)\ **Post date:** [December 30, 2018, 12:47pm UTC](https://community.drivendata.org/t/calculating-the-score/2965/1 "2018-12-30T12:47:29Z")
Hi, How they are calculating the public score ? are they considering about the number of submission as well? Thank you. ---
**Author:** ![utsavporwal31](https://avatars.discourse-cdn.com/v4/letter/u/cab0a1/32.png) [@utsavporwal31](https://community.drivendata.org/u/utsavporwal31)\ **Post date:** [January 26, 2019, 4:27pm UTC](https://community.drivendata.org/t/calculating-the-score/2965/2 "2019-01-26T16:27:50Z")
Its given in the overview of the problem. hey can you tell me what can i do to increase the performance of the model my current score is 0.4512 . Can you give me some tips?? ---
**Author:** ![finalspace](https://avatars.discourse-cdn.com/v4/letter/f/3da27b/32.png) [@finalspace](https://community.drivendata.org/u/finalspace)\ **Post date:** [April 14, 2023, 8:59am UTC](https://community.drivendata.org/t/calculating-the-score/2965/3 "2023-04-14T08:59:48Z")
I am still confused about the leaderboard score. In the [leaderboard results](https://www.drivendata.org/competitions/2/warm-up-predict-blood-donations/leaderboard/) it is mentioned that the log loss is used. In the notebooks published [here](https://github.com/grfiv/predict-blood-donations) the ranking is between 0.4 and 1.8 but the calculated log loss in the notebooks is above 6. The score in the ranking must be something different than the log loss, right?