{ "id": 151505, "title": "W5 LB scorecard & analysis update 0611", "content": "

Please upvote if you think this is usefull to follow, then I know if it's good to continue the updates and not creating lots of spam :) And for you who click the downvote, this is not a state-of-the-art scorecard, please contribute with a new scorecard, you probably do better than me at this, I just making a good try :)

\n

An idea to the scorecard is also to get more use of the analysis and see some patterns that helps to create better models and predictions to those in need e.g. identify models there are better in short, medium and long term, use them thereafter and combine them.
\nExplanation to the columns:
\nScore diff - is the % difference between 2 LB scores(1d = 1 day difference)
\nDist. to the top - distance to the leader.
\nModels - which technique(LGB, math, Fastai etc) that are used in the models, in top 20, can be interesting to have and follow. Because a team has two submission, I just add what I see, and complements ahead to the team's total.
\nDiff from last day - the difference between the distance to the leader, from yesterday, approach the leader or lose momentum.
\nKPI \"1d(day) best mover…\" - Total move(score difference) the group did against the leader and the individual contribution. Track the individual movement within the group against the leader, see the performance in that timeline regardless what speed has as a whole.
\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc -
\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n

One can calculate in many different ways to see that comparison, this was a quick way to do it.

\n

Also check notebooks below for more analysis.
\n https://www.kaggle.com/gaborfodor/w5-top-submissions
\n https://www.kaggle.com/david1013/covid-19-week-5-daily-predictions-top-teams/

", "comments": [ { "id": 882300, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-11T17:46:51.193000", "content": "

That was the final and last update! \nWell done everyone with an important work, especially @gaborfodor who created the #1 model! :)

\n\n

Now we hope that this horrible virus will disappear quickly and cures are created immediately. The thoughts go to everyone affected.

", "votes": 3, "replies": [ { "id": 886181, "author_name": "", "author_url": "", "post_date": "2020-06-14T19:13:01.600000", "content": "

Thank you for your work

\n\n

This was a really interesting competition that I wasn't a part of 👍 👍 💪 💪

", "votes": 1, "replies": [] } ] }, { "id": 882294, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-11T17:40:01.923000", "content": "

Update 0611\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591897120991013&alt=media)\n\"\".png?generation=1591897121767768&alt=media)\n\"\".png?generation=1591897120972891&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591897121662173&alt=media)\nScore distribution all teams from start\n\"\".png?generation=1591897122409789&alt=media)

", "votes": 1, "replies": [] }, { "id": 868399, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-31T07:15:59.930000", "content": "

Update 0530

\n\n

\"\"\n\"\"\n\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top

\n\n

\"\".png?generation=1590909315821566&alt=media)\n\"\".png?generation=1590909313209295&alt=media)\n\"\".png?generation=1590909315421603&alt=media)\nScore and LB Top 20\n\"\".png?generation=1590909313769494&alt=media)

", "votes": 1, "replies": [] }, { "id": 864019, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-27T17:56:31.300000", "content": "

Update 0527

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1590602119748761&alt=media)

\n\n

\"\".png?generation=1590602132315735&alt=media)

\n\n

\"\".png?generation=1590602146614575&alt=media)

\n\n

Score and LB Top 20\n\"\".png?generation=1590602186808334&alt=media)

", "votes": 1, "replies": [ { "id": 864775, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-28T07:18:37.923000", "content": "

Halfway into w5 we can see some patterns. I think my model could predict with more accuracy with NN in the first days. The model also works as one of more ensemble model to balance the predicted trends with the RF average. So far in W5 it looks like an ensemble with the top 3 models from teams DiDoKaPs/disinfectant injection/Lockdown / Northquay would be the best model to use and maybe with a increasing and decreasing ensemble weight, with DiDoKaPs and also add Bernhard Klinger as the Long term models.

", "votes": 0, "replies": [] } ] }, { "id": 861044, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-25T19:46:32.613000", "content": "

Update 0525

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores and Dist to the top

\n\n

\"\".png?generation=1590435822310254&alt=media)

\n\n

\"\".png?generation=1590435836052528&alt=media)

\n\n

\"\".png?generation=1590435854255671&alt=media)

\n\n

-Score and LB Top 20\n\"\".png?generation=1590435898607450&alt=media)

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 11 day period\n11d LB change Top 31-60 & Top 31-60 11d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n\n

\"\"

", "votes": 1, "replies": [] }, { "id": 859826, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-24T19:15:26.570000", "content": "

Update 0524

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores and Dist to the top\n\"\".png?generation=1590347536456116&alt=media)

\n\n

\"\".png?generation=1590347550165971&alt=media)

\n\n

\"\".png?generation=1590347564927746&alt=media)

\n\n

-Score and LB Top 20\n\"\".png?generation=1590347632886065&alt=media)

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 10 day period\n10d LB change Top 31-60 & Top 31-60 10d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n\n

\"\"

", "votes": 1, "replies": [] }, { "id": 856151, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-21T14:26:02.950000", "content": "

Update 0521

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 7 day period\n7d LB change Top 31-60 & Top 31-60 7d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the score gap and Dist to the top gap

\n\n

\"\".png?generation=1590071096866296&alt=media)

\n\n

\"\".png?generation=1590071114157418&alt=media)

\n\n

\"\".png?generation=1590071130310648&alt=media)

\n\n

-Score and LB Top 15\n\"\".png?generation=1590071156855760&alt=media)

", "votes": 1, "replies": [] }, { "id": 855228, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-20T17:27:41.353000", "content": "

Update 0520

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 6 day period\n6d LB change Top 31-60 & Top 31-60 6d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n\n

\"\"

\n\n

Additional scorecard II -\nVisualizing the score gap and Dist to the top gap

\n\n

\"\".png?generation=1589995597625688&alt=media)

\n\n

\"\".png?generation=1589995611904992&alt=media)

\n\n

\"\".png?generation=1589995628159425&alt=media)

\n\n

-Score and LB Top 15\n\"\".png?generation=1589995655144446&alt=media)

", "votes": 1, "replies": [ { "id": 855277, "author_name": "", "author_url": "", "post_date": "2020-05-20T18:18:54.110000", "content": "

Good job @kirderf! Moving up strong +1

", "votes": 0, "replies": [] }, { "id": 855302, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-20T19:03:08.973000", "content": "

Thanks! :) But amazing how well the top teams are performing, but not quite surprising david1013 sasrdw aerdem4 philippsinger dott1718 kazanova christofhenkel , all the top grandmaster? Besides cpmpml whom we miss.

", "votes": 1, "replies": [] }, { "id": 855534, "author_name": "grayjay", "author_url": "", "post_date": "2020-05-21T02:43:24.603000", "content": "

Grayjay tumbles out of top ten !

\n\n

Note to self -- a one-line autoregressive equation for p50 might be ok, but for the p05 and p95 boundaries it's a really terrible idea. Oops.

\n\n

Actually, I'm amazed this model did so well. It was a \"simplest thing that might possibly work\"; less than 50 lines of code. Eight days in the top ten is a long time in the sun for something that primitive.

\n\n

Best of luck to everyone...

", "votes": 2, "replies": [] }, { "id": 855842, "author_name": "Dieter", "author_url": "", "post_date": "2020-05-21T09:12:19.110000", "content": "

@kirderf thank you for pinging me (and all others). Otherwise I might have missed this very important post.

", "votes": 0, "replies": [] }, { "id": 855859, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-21T09:28:46.843000", "content": "

Important and important just a positive feedback on your models performance, good work with them :)

", "votes": 0, "replies": [] }, { "id": 855886, "author_name": "CPMP", "author_url": "", "post_date": "2020-05-21T10:03:50.020000", "content": "

@kirderf thanks, rooting for you in this competition!

\n\n

I am not sure I'm missed by everyone to be honest ;)

", "votes": 1, "replies": [] }, { "id": 855963, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-21T11:20:39.683000", "content": "

Thanks @cpmpml! Hope we will see lots of teams again, more teams, better development, more help to those in need.

", "votes": 0, "replies": [] }, { "id": 855967, "author_name": "Ahmet Erdem", "author_url": "", "post_date": "2020-05-21T11:24:37.620000", "content": "

I hope we don't see W6.

", "votes": 2, "replies": [] }, { "id": 855982, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-21T11:41:40.413000", "content": "

I see your point, I think you saw mine, ment if we will see a W6 and I hope we see lots of teams again to the model dev, important contribution

", "votes": 0, "replies": [] } ] }, { "id": 853007, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-18T22:09:33.280000", "content": "

Update 0518

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 4 day period\n4d LB change Top 31-60 & Top 31-60 4d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev4dscore)/4))/2)PredDayAhead

\n\n

\"\"

", "votes": 1, "replies": [ { "id": 853043, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-18T23:27:40.847000", "content": "

After 4 days we can see that it's really even between the models in top 5 and in top 10 @bernardklinger has a good momentum followed by @neg0000273, great work!

", "votes": 0, "replies": [] }, { "id": 853616, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-19T10:36:07.693000", "content": "

Additional charts

\n\n

\"\".png?generation=1589884516042426&alt=media)

\n\n

\"\".png?generation=1589884536367951&alt=media)

\n\n

\"\"

", "votes": 1, "replies": [] }, { "id": 853659, "author_name": "Russ Wolfinger", "author_url": "", "post_date": "2020-05-19T11:20:52.820000", "content": "

I was surprised by the decreasing scores over the two weekend days, but expect they will shoot up again now that we are into the work week.

", "votes": 2, "replies": [] }, { "id": 853663, "author_name": "Ahmet Erdem", "author_url": "", "post_date": "2020-05-19T11:25:27.810000", "content": "

If there was no peak in French fatalities, scores would be even less.

", "votes": 1, "replies": [] }, { "id": 853763, "author_name": "Psi", "author_url": "", "post_date": "2020-05-19T13:15:52.027000", "content": "

Apparently I dont understand the metric because I was also thinking it should go up :D

", "votes": 1, "replies": [] }, { "id": 853830, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-19T14:18:30.237000", "content": "

Additional charts II \nScore and LB Top 15

\n\n

\"\".png?generation=1589897899410364&alt=media)

", "votes": 1, "replies": [] } ] }, { "id": 851647, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-17T19:30:34.033000", "content": "

Update 0517

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 3 day period\n3d LB change Top 31-60 & Top 31-60 3d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev3dscore)/3))/2)PredDayAhead

\n\n

\"\"

", "votes": 1, "replies": [ { "id": 851796, "author_name": "Giba", "author_url": "", "post_date": "2020-05-17T23:38:23.953000", "content": "

Hi @kirderf, check the notebook, my model is QuantReg from statsmodels and not XGB.

", "votes": 0, "replies": [] }, { "id": 851836, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-18T00:41:57.690000", "content": "

@titericz perfect thanks, your notebooks I wasn't 100%, I just saw a private dataset in the beginning and imports of\nfrom sklearn.ensemble import GradientBoostingRegressor\nfrom sklearn.ensemble import GradientBoostingRegressor\nfrom lightgbm import LGBMRegressor\nimport xgboost as xgb.\nThought the dataset could have had pretrained models.\nYou will see the update tomorrow :)

", "votes": 0, "replies": [] }, { "id": 852823, "author_name": "Giba", "author_url": "", "post_date": "2020-05-18T17:47:40.523000", "content": "

I believe my notebook needed some garbage collector. I tried many algorithms but a simple QuantReg performed best.

", "votes": 2, "replies": [] }, { "id": 853038, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-18T23:16:05.477000", "content": "

Yes it's hard to predict and follow the short trends in top/bottoms, average works better.

", "votes": 0, "replies": [] } ] }, { "id": 853937, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-19T16:09:54.933000", "content": "

Update 0519

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 5 day period\n5d LB change Top 31-60 & Top 31-60 5d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n\n

\"\"

\n\n

Additional scorecard II -\n- Visualizing the score gap and Dist to the top gap\n\"\".png?generation=1589904552714636&alt=media)

\n\n

\"\".png?generation=1589904570030207&alt=media)

\n\n

\"\".png?generation=1589904682286367&alt=media)

\n\n

-Score and LB Top 15\n\"\".png?generation=1589904586617118&alt=media)

", "votes": 2, "replies": [ { "id": 854022, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-19T17:20:15.677000", "content": "

Amazing how the top 4 models are in sync with each other, and almost no difference in scores between the days, good work!\nGreat work again with your model @moradnejad, this week maybe even better :)

\n\n

Look at the \"Score gap top 50\" line chart above, some models in the bottom 50 reduce their values ​​quickly, must study what algos they use.

", "votes": 1, "replies": [] }, { "id": 854100, "author_name": "", "author_url": "", "post_date": "2020-05-19T19:00:34.730000", "content": "

Wow, a great day for me!

\n\n

Also, very great visualizations for tracking changes 💪 @kirderf

", "votes": 0, "replies": [] } ] }, { "id": 850434, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-16T16:17:07.337000", "content": "

Update 0516

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 2 day period\n2d LB change Top 31-60 & Top 31-60 2d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\n((TodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev2dscore)/2))/2)PredDayAhead)+TodayScore(((1dAveScoreDiff+2dAveScoreDiff)/2)+1)^PredDayAhead)/2

\n\n

\"\"

", "votes": 2, "replies": [ { "id": 850478, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-16T17:18:43.317000", "content": "

my model seems to have some momentum :) also @bernhardklinger and @rohanrao after 2 days.

", "votes": 2, "replies": [] }, { "id": 851622, "author_name": "", "author_url": "", "post_date": "2020-05-17T19:01:49.417000", "content": "

Thanks for doing this again. It's fun to follow +1

", "votes": 0, "replies": [] }, { "id": 851756, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-17T22:15:34.770000", "content": "

Thanks :) I think it's a good way to keep track of all movements between the models from top to bottom, identify patterns. Each team can also compare more easy and do thier own analysis based on the idea/architecture of the model.

", "votes": 1, "replies": [] } ] }, { "id": 849432, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-15T18:43:37.607000", "content": "

Update 0515

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 1 day period\n22d LB change Top 31-60 & Top 31-60 1d ave. % score diff.\nPrediction Top20 1 day ahead predictions, use actual score value change added to the new #, times days.

\n\n

\"\"

", "votes": 2, "replies": [ { "id": 849439, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-15T18:51:02.220000", "content": "

I've added which technique(LGB, math, Fastai etc) that are used in the models, in top 20, can be interesting to have and follow. Because a team has two submission, I just add what I see, and complements ahead to the team's total.

", "votes": 2, "replies": [] }, { "id": 849836, "author_name": "Ahmet Erdem", "author_url": "", "post_date": "2020-05-16T05:33:44.263000", "content": "

Nice that you have added which model is used for each!

\n\n

Btw you can also create Day LBs by calculating n*score[n] - (n-1)*score[n-1]

", "votes": 2, "replies": [] }, { "id": 849939, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-16T07:34:31.553000", "content": "

Thanks! :) I'll take that in count. I think I will try a mix of prev scorecard pred calc and with some modifications, to use in this scorecard's pred columns, for 1, 3 and 5 days pred ahead.\n((TodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead)+TodayScore(((1dAveScoreDiff+5dAveScoreDiff)/2)+1)^PredDayAhead)/2

", "votes": 0, "replies": [] }, { "id": 850582, "author_name": "Vopani", "author_url": "", "post_date": "2020-05-16T19:38:15.747000", "content": "

@kirderf Thanks for maintaining this. I haven't had a chance to look at other solutions, but I was just wondering aren't there any ensemble models being used?

", "votes": 1, "replies": [] }, { "id": 851018, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-17T09:23:40.147000", "content": "

@rohanrao thanks, I'll continue as long there is an interest for it :) No not many what I can see, they with larger teams look to ensemble each other's models. Otherwise most weighted average ensembles of the same model / ML.

", "votes": 0, "replies": [] } ] }, { "id": 881295, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-10T22:24:00.420000", "content": "

Update 0610\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591827776863571&alt=media)\n\"\".png?generation=1591827775838966&alt=media)\n\"\".png?generation=1591827775822026&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591827817311542&alt=media)\nScore distribution all teams from start\n\"\".png?generation=1591827835180455&alt=media)

", "votes": 0, "replies": [] }, { "id": 879830, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-09T18:48:07.927000", "content": "

Update 0609\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591728373043015&alt=media)\n\"\".png?generation=1591728372916590&alt=media)\n\"\".png?generation=1591728375088400&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591728457203778&alt=media)\nScore distribution all teams from start\n\"\".png?generation=1591728484912030&alt=media)

", "votes": 0, "replies": [] }, { "id": 878852, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-08T23:35:27.980000", "content": "

Update 0608\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591659267032290&alt=media)\n\"\".png?generation=1591659265455708&alt=media)\n\"\".png?generation=1591659266694443&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591659298022027&alt=media)\nScore distribution all teams from start\n\"\".png?generation=1591659322345161&alt=media)

", "votes": 0, "replies": [] }, { "id": 877741, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-07T23:01:00.953000", "content": "

Update 0607\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591570712872617&alt=media)\n\"\".png?generation=1591570711698441&alt=media)\n\"\".png?generation=1591570711094114&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591570753800488&alt=media)\nScore distribution all teams from start\n\"\".png?generation=1591570856410573&alt=media)

", "votes": 0, "replies": [] }, { "id": 876429, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-06T17:17:08.647000", "content": "

Update 0606\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591464311422864&alt=media)\n\"\".png?generation=1591463760243147&alt=media)\n\"\".png?generation=1591463761005217&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591463808578363&alt=media)

", "votes": 0, "replies": [] }, { "id": 875352, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-05T17:31:53.317000", "content": "

Update 0605

\n\n

\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591378269792155&alt=media)\n\"\".png?generation=1591378269265438&alt=media)\n\"\".png?generation=1591378271765835&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591378307733273&alt=media)

", "votes": 0, "replies": [] }, { "id": 873901, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-04T14:02:54.493000", "content": "

Update 0604\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591279348929407&alt=media)\n\"\".png?generation=1591279349624742&alt=media)\n\"\".png?generation=1591279349144642&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591279349469668&alt=media)

", "votes": 0, "replies": [] }, { "id": 872970, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-03T16:50:46.773000", "content": "

Update 0603\n\"\"\n\"\"\n\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591203009300088&alt=media)\n\"\".png?generation=1591203009726542&alt=media)\n\"\".png?generation=1591203009074868&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591203009488129&alt=media)

", "votes": 0, "replies": [ { "id": 873034, "author_name": "Psi", "author_url": "", "post_date": "2020-06-03T17:57:34.733000", "content": "

What happened to the first chart? :)

\n\n

Thanks for posting this regularly!

", "votes": 1, "replies": [] }, { "id": 873051, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-03T18:18:33.420000", "content": "

It will come soon :)

", "votes": 1, "replies": [] } ] }, { "id": 871703, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-02T15:24:34.703000", "content": "

Update 0602\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591111418284109&alt=media)\n\"\".png?generation=1591111418692065&alt=media)\n\"\".png?generation=1591111420200103&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591111418137060&alt=media)

", "votes": 0, "replies": [] }, { "id": 870519, "author_name": "Kirderf", "author_url": "", "post_date": "2020-06-01T18:05:10.733000", "content": "

Update 0601

\n\n

\"\"\n\"\"\n\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591034761254106&alt=media)\n\"\".png?generation=1591034757575379&alt=media)\n\"\".png?generation=1591034759622582&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591034796853624&alt=media)

", "votes": 0, "replies": [] }, { "id": 869164, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-31T19:28:25.073000", "content": "

Update 0531

\n\n

\"\"\n\"\"\n\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top

\n\n

\"\".png?generation=1590953275475090&alt=media)\n\"\".png?generation=1590953276248377&alt=media)\n\"\".png?generation=1590953275777142&alt=media)\nScore and LB Top 20\n\"\".png?generation=1590953275909618&alt=media)

", "votes": 0, "replies": [] }, { "id": 866819, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-29T17:48:11.207000", "content": "

Update 0529

\n\n

\"\"\n\"\"\n\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1590774460154868&alt=media)\n\"\".png?generation=1590774460125443&alt=media)\n\"\".png?generation=1590774460457662&alt=media)\nScore and LB Top 20\n\"\".png?generation=1590774460161138&alt=media)

", "votes": 0, "replies": [] }, { "id": 865533, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-28T17:24:42.230000", "content": "

Update 0528

\n\n

Halfway into w5 we can see some patterns. So far in W5 it looks like an ensemble with the top 3 models from teams DiDoKaPs/disinfectant injection/Lockdown / Northquay would be the model to use and maybe with an increasing and decreasing ensemble weight, with DiDoKaPs and also add Bernhard Klinger and Beluga as Long term models.\nI think my own model could predict with more accuracy using NN the first days. The model may also work as one of more ensemble models to balance the predicted trends with the RF average.

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top

\n\n

\"\".png?generation=1590686488627164&alt=media)\n\"\".png?generation=1590686505247899&alt=media)\n\"\".png?generation=1590686505659542&alt=media)\nScore and LB Top 20\n\"\".png?generation=1590686504696759&alt=media)

", "votes": 0, "replies": [] }, { "id": 862539, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-26T16:42:49.243000", "content": "

Update 0526

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top

\n\n

\"\".png?generation=1590511320541250&alt=media)

\n\n

\"\".png?generation=1590511337419125&alt=media)

\n\n

\"\".png?generation=1590511352099860&alt=media)

\n\n

Score and LB Top 20\n\"\".png?generation=1590511366180353&alt=media)

", "votes": 0, "replies": [] }, { "id": 858722, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-23T18:22:32.020000", "content": "

Update 0523

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores and Dist to the top

\n\n

\"\".png?generation=1590258089536013&alt=media)

\n\n

\"\".png?generation=1590258106365435&alt=media)

\n\n

\"\".png?generation=1590258120804038&alt=media)

\n\n

-Score and LB Top 20\n\"\".png?generation=1590258145933555&alt=media)

", "votes": 0, "replies": [] }, { "id": 858313, "author_name": "Kirderf", "author_url": "", "post_date": "2020-05-23T11:19:00.393000", "content": "

Update 0522 - reduced number of charts due to 38 teams missing in the top 50-88 in yesterday's LB

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the score gap and Dist to the top gap\n\"\".png?generation=1590232636721717&alt=media)

\n\n

\"\".png?generation=1590232650764779&alt=media)

\n\n

\"\".png?generation=1590232664751648&alt=media)

\n\n

-Score and LB Top 20\n\"\".png?generation=1590232697768089&alt=media)

", "votes": 0, "replies": [] }, { "id": 857523, "author_name": "", "author_url": "", "post_date": "2020-05-22T17:17:08.657000", "content": "", "votes": 0, "replies": [] } ], "comment_count": 63, "messages": [ { "id": 849427, "postDate": "2020-05-15T18:38:21.853Z", "content": "

Please upvote if you think this is usefull to follow, then I know if it's good to continue the updates and not creating lots of spam :) And for you who click the downvote, this is not a state-of-the-art scorecard, please contribute with a new scorecard, you probably do better than me at this, I just making a good try :)

\n

An idea to the scorecard is also to get more use of the analysis and see some patterns that helps to create better models and predictions to those in need e.g. identify models there are better in short, medium and long term, use them thereafter and combine them.
\nExplanation to the columns:
\nScore diff - is the % difference between 2 LB scores(1d = 1 day difference)
\nDist. to the top - distance to the leader.
\nModels - which technique(LGB, math, Fastai etc) that are used in the models, in top 20, can be interesting to have and follow. Because a team has two submission, I just add what I see, and complements ahead to the team's total.
\nDiff from last day - the difference between the distance to the leader, from yesterday, approach the leader or lose momentum.
\nKPI \"1d(day) best mover…\" - Total move(score difference) the group did against the leader and the individual contribution. Track the individual movement within the group against the leader, see the performance in that timeline regardless what speed has as a whole.
\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc -
\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n

One can calculate in many different ways to see that comparison, this was a quick way to do it.

\n

Also check notebooks below for more analysis.
\n https://www.kaggle.com/gaborfodor/w5-top-submissions
\n https://www.kaggle.com/david1013/covid-19-week-5-daily-predictions-top-teams/

", "rawMarkdown": "Please upvote if you think this is usefull to follow, then I know if it's good to continue the updates and not creating lots of spam :) And for you who click the downvote, this is not a state-of-the-art scorecard, please contribute with a new scorecard, you probably do better than me at this, I just making a good try :)\n\nAn idea to the scorecard is also to get more use of the analysis and see some patterns that helps to create better models and predictions to those in need e.g. identify models there are better in short, medium and long term, use them thereafter and combine them.\nExplanation to the columns:\nScore diff - is the % difference between 2 LB scores(1d = 1 day difference)\nDist. to the top - distance to the leader.\nModels - which technique(LGB, math, Fastai etc) that are used in the models, in top 20, can be interesting to have and follow. Because a team has two submission, I just add what I see, and complements ahead to the team's total.\nDiff from last day - the difference between the distance to the leader, from yesterday, approach the leader or lose momentum.\nKPI \"1d(day) best mover…\" - Total move(score difference) the group did against the leader and the individual contribution. Track the individual movement within the group against the leader, see the performance in that timeline regardless what speed has as a whole.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc - \nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead\n\nOne can calculate in many different ways to see that comparison, this was a quick way to do it.\n\nAlso check notebooks below for more analysis.\n https://www.kaggle.com/gaborfodor/w5-top-submissions \n https://www.kaggle.com/david1013/covid-19-week-5-daily-predictions-top-teams/\n", "votes": 22 }, { "id": 882300, "postDate": "2020-06-11T17:46:51.193Z", "content": "

That was the final and last update! \nWell done everyone with an important work, especially @gaborfodor who created the #1 model! :)

\n\n

Now we hope that this horrible virus will disappear quickly and cures are created immediately. The thoughts go to everyone affected.

", "rawMarkdown": "That was the final and last update! \nWell done everyone with an important work, especially @gaborfodor who created the #1 model! :)\n\nNow we hope that this horrible virus will disappear quickly and cures are created immediately. The thoughts go to everyone affected.", "votes": 3, "replies": [ { "id": 886181, "postDate": "2020-06-14T19:13:01.600Z", "content": "

Thank you for your work

\n\n

This was a really interesting competition that I wasn't a part of 👍 👍 💪 💪

", "rawMarkdown": "Thank you for your work\n\nThis was a really interesting competition that I wasn't a part of 👍 👍 💪 💪 ", "votes": 1 } ] }, { "id": 882294, "postDate": "2020-06-11T17:40:01.923Z", "content": "

Update 0611\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591897120991013&alt=media)\n\"\".png?generation=1591897121767768&alt=media)\n\"\".png?generation=1591897120972891&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591897121662173&alt=media)\nScore distribution all teams from start\n\"\".png?generation=1591897122409789&alt=media)

", "rawMarkdown": "Update 0611\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F75df848bc38c12d1f15abf5611f24ef2%2F611_1.jpg?generation=1591897088370287&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fbed4633a3ee32c0d0f3a0a1afa5361e5%2F611_2.png?generation=1591897088353323&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Ffd038e65ee269059e85a355a87e34299%2F611_3.png?generation=1591897089015973&alt=media)\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F84fef648b05fc87ccd08b5742a13d3b4%2Fnedladdning%20(58).png?generation=1591897120991013&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F1a2711bd38a59c03cf174cff6cf74eff%2Fnedladdning%20(59).png?generation=1591897121767768&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F97c5719431646ae069ebed4c3d723af3%2Fnedladdning%20(60).png?generation=1591897120972891&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F2b40b725fa431241d6d48ae896166e73%2Fnedladdning%20(61).png?generation=1591897121662173&alt=media)\nScore distribution all teams from start\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fecc8bc90c616abc489ec78cda0366a8c%2Fnedladdning%20(62).png?generation=1591897122409789&alt=media)", "votes": 1 }, { "id": 868399, "postDate": "2020-05-31T07:15:59.930Z", "content": "

Update 0530

\n\n

\"\"\n\"\"\n\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top

\n\n

\"\".png?generation=1590909315821566&alt=media)\n\"\".png?generation=1590909313209295&alt=media)\n\"\".png?generation=1590909315421603&alt=media)\nScore and LB Top 20\n\"\".png?generation=1590909313769494&alt=media)

", "rawMarkdown": "Update 0530\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F45820d69a80b74874b7bde712ad22fb4%2F530_1.jpg?generation=1590909265209926&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F18dd852286233b440e9ee67802f35203%2F530_2.png?generation=1590909264637739&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F30015c2a1ee6a855aa5ec744086ef58a%2F530_3.png?generation=1590909265930537&alt=media)\n\nAdditional charts -\nVisualizing the scores & Dist to the top\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F594b0b03dd2997c5afa4b3307b7fc879%2Fnedladdning%20(48).png?generation=1590909315821566&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F3c9e424c9177a57f554c449a785fb46c%2Fnedladdning%20(49).png?generation=1590909313209295&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fddbf4fe2d979887be127bad03ed3d775%2Fnedladdning%20(50).png?generation=1590909315421603&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Feb7ae4d8e7efafa958c96809e6d168d4%2Fnedladdning%20(51).png?generation=1590909313769494&alt=media)", "votes": 1 }, { "id": 864019, "postDate": "2020-05-27T17:56:31.300Z", "content": "

Update 0527

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1590602119748761&alt=media)

\n\n

\"\".png?generation=1590602132315735&alt=media)

\n\n

\"\".png?generation=1590602146614575&alt=media)

\n\n

Score and LB Top 20\n\"\".png?generation=1590602186808334&alt=media)

", "rawMarkdown": "Update 0527\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fc43a822af53d08235fb350e2d84e5642%2F527_1.jpg?generation=1590602927264613&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F3cb48b76a0f178945eff5a79415886d6%2F527_2.png?generation=1590602943333768&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F858af213fbe3c737f885add511663965%2F527_3.png?generation=1590602091760044&alt=media)\n\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F1fd2b5e9c16e4d04e53d8d10cddaabaa%2Fnedladdning%20(36).png?generation=1590602119748761&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F5890d15cd1360d20ecf405f3ab2ba750%2Fnedladdning%20(37).png?generation=1590602132315735&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F0041e80675046ab528b92796c945036c%2Fnedladdning%20(38).png?generation=1590602146614575&alt=media)\n\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F9a3f7b40cf89838e90ac2a05c9a72be1%2Fnedladdning%20(39).png?generation=1590602186808334&alt=media)", "votes": 1, "replies": [ { "id": 864775, "postDate": "2020-05-28T07:18:37.923Z", "content": "

Halfway into w5 we can see some patterns. I think my model could predict with more accuracy with NN in the first days. The model also works as one of more ensemble model to balance the predicted trends with the RF average. So far in W5 it looks like an ensemble with the top 3 models from teams DiDoKaPs/disinfectant injection/Lockdown / Northquay would be the best model to use and maybe with a increasing and decreasing ensemble weight, with DiDoKaPs and also add Bernhard Klinger as the Long term models.

", "rawMarkdown": "Halfway into w5 we can see some patterns. I think my model could predict with more accuracy with NN in the first days. The model also works as one of more ensemble model to balance the predicted trends with the RF average. So far in W5 it looks like an ensemble with the top 3 models from teams DiDoKaPs/disinfectant injection/Lockdown / Northquay would be the best model to use and maybe with a increasing and decreasing ensemble weight, with DiDoKaPs and also add Bernhard Klinger as the Long term models." } ] }, { "id": 861044, "postDate": "2020-05-25T19:46:32.613Z", "content": "

Update 0525

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores and Dist to the top

\n\n

\"\".png?generation=1590435822310254&alt=media)

\n\n

\"\".png?generation=1590435836052528&alt=media)

\n\n

\"\".png?generation=1590435854255671&alt=media)

\n\n

-Score and LB Top 20\n\"\".png?generation=1590435898607450&alt=media)

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 11 day period\n11d LB change Top 31-60 & Top 31-60 11d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n\n

\"\"

", "rawMarkdown": "Update 0525\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F9450b320f5fd2d9684049a982618eb96%2F525_1.jpg?generation=1590435512251646&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fd1aa6a1cafa6dded7689a0c16764e412%2F525_2.png?generation=1590435528126325&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F4abe893dd10e65d5a1c9ab4d47cc706d%2F525_3.png?generation=1590435544023317&alt=media)\n\nAdditional charts -\nVisualizing the scores and Dist to the top\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F7cdbb240ce3286e852b93cd378604842%2Fnedladdning%20(28).png?generation=1590435822310254&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F6e9f2102386187e8832b9b052c92699b%2Fnedladdning%20(29).png?generation=1590435836052528&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fa8844d9191edd9ca89ab525add2ab675%2Fnedladdning%20(30).png?generation=1590435854255671&alt=media)\n\n-Score and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fa8e65528b3e58ea1fe095da58d1674ab%2Fnedladdning%20(31).png?generation=1590435898607450&alt=media)\n\nAdditional scorecard -\nBest LB change Top 31-88 - 11 day period\n11d LB change Top 31-60 & Top 31-60 11d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F000381621c9e42878d95781dc037e350%2F525_4.jpg?generation=1590435987082963&alt=media)", "votes": 1 }, { "id": 859826, "postDate": "2020-05-24T19:15:26.570Z", "content": "

Update 0524

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores and Dist to the top\n\"\".png?generation=1590347536456116&alt=media)

\n\n

\"\".png?generation=1590347550165971&alt=media)

\n\n

\"\".png?generation=1590347564927746&alt=media)

\n\n

-Score and LB Top 20\n\"\".png?generation=1590347632886065&alt=media)

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 10 day period\n10d LB change Top 31-60 & Top 31-60 10d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n\n

\"\"

", "rawMarkdown": "Update 0524\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F5d210864f615f0f78a71420919259138%2F0524_1.jpg?generation=1590347478315723&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fca59eda95e9732219a1fff4dc3e521e0%2F0524_2.png?generation=1590347493613671&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Feaea8cc5968240b9077c65cc2db6a44e%2F0524_3.png?generation=1590347507583192&alt=media)\n\nAdditional charts -\nVisualizing the scores and Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fc48a947190d4143c770aaa49c2f2b1e8%2Fnedladdning%20(24).png?generation=1590347536456116&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F4f19f87cc3280292d7fed600481032a6%2Fnedladdning%20(25).png?generation=1590347550165971&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fc88ce4f221097c60ba15d509ca30ccfb%2Fnedladdning%20(26).png?generation=1590347564927746&alt=media)\n\n-Score and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F4719a9d27c85465acc8e87b25ebee6e1%2Fnedladdning%20(27).png?generation=1590347632886065&alt=media)\n\nAdditional scorecard -\nBest LB change Top 31-88 - 10 day period\n10d LB change Top 31-60 & Top 31-60 10d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F6283f594ea961c992ab40548ac618db3%2F0524_4.jpg?generation=1590347693590632&alt=media)\n", "votes": 1 }, { "id": 856151, "postDate": "2020-05-21T14:26:02.950Z", "content": "

Update 0521

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 7 day period\n7d LB change Top 31-60 & Top 31-60 7d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the score gap and Dist to the top gap

\n\n

\"\".png?generation=1590071096866296&alt=media)

\n\n

\"\".png?generation=1590071114157418&alt=media)

\n\n

\"\".png?generation=1590071130310648&alt=media)

\n\n

-Score and LB Top 15\n\"\".png?generation=1590071156855760&alt=media)

", "rawMarkdown": "Update 0521\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F2d650ca2cf5bec29838c1f1f965b531b%2F0521_1.jpg?generation=1590070623274099&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F3dccb1fbaa18ce121ce37be03dd592d5%2F0521_2.png?generation=1590070637992639&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F01ae3bf9ef11a5a2add7fc8373700e10%2F0521_3.png?generation=1590070650792254&alt=media)\n\nAdditional scorecard -\nBest LB change Top 31-88 - 7 day period\n7d LB change Top 31-60 & Top 31-60 7d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F0573f5bda00abe1e68c8c281ef38473e%2F0521_4.jpg?generation=1590070706010424&alt=media)\n\nAdditional charts -\nVisualizing the score gap and Dist to the top gap\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F206153adbf2895660c18a877eae0f4e8%2Fnedladdning%20(12).png?generation=1590071096866296&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F8e03eefa83fa532857479a0de13e7700%2Fnedladdning%20(13).png?generation=1590071114157418&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F459ad266fb81e0c80e5cc38313284f34%2Fnedladdning%20(14).png?generation=1590071130310648&alt=media)\n\n-Score and LB Top 15\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F66485649edddddbfb5f40d18b554db76%2Fnedladdning%20(15).png?generation=1590071156855760&alt=media)", "votes": 1 }, { "id": 855228, "postDate": "2020-05-20T17:27:41.353Z", "content": "

Update 0520

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 6 day period\n6d LB change Top 31-60 & Top 31-60 6d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n\n

\"\"

\n\n

Additional scorecard II -\nVisualizing the score gap and Dist to the top gap

\n\n

\"\".png?generation=1589995597625688&alt=media)

\n\n

\"\".png?generation=1589995611904992&alt=media)

\n\n

\"\".png?generation=1589995628159425&alt=media)

\n\n

-Score and LB Top 15\n\"\".png?generation=1589995655144446&alt=media)

", "rawMarkdown": "Update 0520\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F36db30c01bcac0370a1cc0b7c568d30f%2F0520_1.jpg?generation=1589995497971429&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F20a34f8f6ff63895934b604d06e9a1e0%2F0520_2.png?generation=1589995511511685&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F7d442f04f7c11e88698bd1443704dd8a%2F0520_3.png?generation=1589995525062115&alt=media)\n\nAdditional scorecard -\nBest LB change Top 31-88 - 6 day period\n6d LB change Top 31-60 & Top 31-60 6d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F58f8f94b54990c2dfcf9178a3ca30387%2F0520_4.jpg?generation=1589995560878618&alt=media)\n\nAdditional scorecard II -\nVisualizing the score gap and Dist to the top gap\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fe7c41c73935bbd4925db7332c31ab02d%2Fnedladdning%20(8).png?generation=1589995597625688&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F8bd0a2911c913c1d5abbdc71d23010ce%2Fnedladdning%20(9).png?generation=1589995611904992&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F0424fcde0081fb49ff8ada09ae21235c%2Fnedladdning%20(10).png?generation=1589995628159425&alt=media)\n\n-Score and LB Top 15\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fc41659ad044d68ef636d3cfddeb13113%2Fnedladdning%20(11).png?generation=1589995655144446&alt=media)", "votes": 1, "replies": [ { "id": 855277, "postDate": "2020-05-20T18:18:54.110Z", "content": "

Good job @kirderf! Moving up strong +1

", "rawMarkdown": "Good job @kirderf! Moving up strong +1" }, { "id": 855302, "postDate": "2020-05-20T19:03:08.973Z", "content": "

Thanks! :) But amazing how well the top teams are performing, but not quite surprising david1013 sasrdw aerdem4 philippsinger dott1718 kazanova christofhenkel , all the top grandmaster? Besides cpmpml whom we miss.

", "rawMarkdown": " Thanks! :) But amazing how well the top teams are performing, but not quite surprising david1013 sasrdw aerdem4 philippsinger dott1718 kazanova christofhenkel , all the top grandmaster? Besides cpmpml whom we miss.", "votes": 1 }, { "id": 855534, "postDate": "2020-05-21T02:43:24.603Z", "content": "

Grayjay tumbles out of top ten !

\n\n

Note to self -- a one-line autoregressive equation for p50 might be ok, but for the p05 and p95 boundaries it's a really terrible idea. Oops.

\n\n

Actually, I'm amazed this model did so well. It was a \"simplest thing that might possibly work\"; less than 50 lines of code. Eight days in the top ten is a long time in the sun for something that primitive.

\n\n

Best of luck to everyone...

", "rawMarkdown": "Grayjay tumbles out of top ten !\n\nNote to self -- a one-line autoregressive equation for p50 might be ok, but for the p05 and p95 boundaries it's a really terrible idea. Oops.\n\nActually, I'm amazed this model did so well. It was a \"simplest thing that might possibly work\"; less than 50 lines of code. Eight days in the top ten is a long time in the sun for something that primitive.\n\nBest of luck to everyone...", "votes": 2 }, { "id": 855842, "postDate": "2020-05-21T09:12:19.110Z", "content": "

@kirderf thank you for pinging me (and all others). Otherwise I might have missed this very important post.

", "rawMarkdown": "@kirderf thank you for pinging me (and all others). Otherwise I might have missed this very important post." }, { "id": 855859, "postDate": "2020-05-21T09:28:46.843Z", "content": "

Important and important just a positive feedback on your models performance, good work with them :)

", "rawMarkdown": "Important and important just a positive feedback on your models performance, good work with them :)" }, { "id": 855886, "postDate": "2020-05-21T10:03:50.020Z", "content": "

@kirderf thanks, rooting for you in this competition!

\n\n

I am not sure I'm missed by everyone to be honest ;)

", "rawMarkdown": "@kirderf thanks, rooting for you in this competition! \n\nI am not sure I'm missed by everyone to be honest ;)", "votes": 1 }, { "id": 855963, "postDate": "2020-05-21T11:20:39.683Z", "content": "

Thanks @cpmpml! Hope we will see lots of teams again, more teams, better development, more help to those in need.

", "rawMarkdown": "Thanks @cpmpml! Hope we will see lots of teams again, more teams, better development, more help to those in need." }, { "id": 855967, "postDate": "2020-05-21T11:24:37.620Z", "content": "

I hope we don't see W6.

", "rawMarkdown": "I hope we don't see W6.", "votes": 2 }, { "id": 855982, "postDate": "2020-05-21T11:41:40.413Z", "content": "

I see your point, I think you saw mine, ment if we will see a W6 and I hope we see lots of teams again to the model dev, important contribution

", "rawMarkdown": "I see your point, I think you saw mine, ment if we will see a W6 and I hope we see lots of teams again to the model dev, important contribution" } ] }, { "id": 853007, "postDate": "2020-05-18T22:09:33.280Z", "content": "

Update 0518

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 4 day period\n4d LB change Top 31-60 & Top 31-60 4d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev4dscore)/4))/2)PredDayAhead

\n\n

\"\"

", "rawMarkdown": "Update 0518\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F5579f0768153bfb782ee8e150e918305%2F0518_1.jpg?generation=1589839737617035&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F819b1eee1bb1f898c9432806ecc89472%2F0518_2.png?generation=1589839517348798&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F379443aaac68afc27e7eab6289864641%2F0518_3.png?generation=1589839534275443&alt=media)\n\nAdditional scorecard -\nBest LB change Top 31-88 - 4 day period\n4d LB change Top 31-60 & Top 31-60 4d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev4dscore)/4))/2)PredDayAhead\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fbb086fdeeb238d762eedff2fe51ca764%2F0518_4.jpg?generation=1589839561810161&alt=media)", "votes": 1, "replies": [ { "id": 853043, "postDate": "2020-05-18T23:27:40.847Z", "content": "

After 4 days we can see that it's really even between the models in top 5 and in top 10 @bernardklinger has a good momentum followed by @neg0000273, great work!

", "rawMarkdown": "After 4 days we can see that it's really even between the models in top 5 and in top 10 @bernardklinger has a good momentum followed by @neg0000273, great work!" }, { "id": 853616, "postDate": "2020-05-19T10:36:07.693Z", "content": "

Additional charts

\n\n

\"\".png?generation=1589884516042426&alt=media)

\n\n

\"\".png?generation=1589884536367951&alt=media)

\n\n

\"\"

", "rawMarkdown": "Additional charts\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fd37afe851de30b43eb4dfbaf4b0b087f%2Fnedladdning%20(2).png?generation=1589884516042426&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Ff51c54320d4856a8ed1d9c5a9d860062%2Fnedladdning%20(1).png?generation=1589884536367951&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F58d15f1b8b01c51661303350b670ee14%2Fnedladdning.png?generation=1589884548359851&alt=media)\n", "votes": 1 }, { "id": 853659, "postDate": "2020-05-19T11:20:52.820Z", "content": "

I was surprised by the decreasing scores over the two weekend days, but expect they will shoot up again now that we are into the work week.

", "rawMarkdown": "I was surprised by the decreasing scores over the two weekend days, but expect they will shoot up again now that we are into the work week.", "votes": 2 }, { "id": 853663, "postDate": "2020-05-19T11:25:27.810Z", "content": "

If there was no peak in French fatalities, scores would be even less.

", "rawMarkdown": "If there was no peak in French fatalities, scores would be even less.", "votes": 1 }, { "id": 853763, "postDate": "2020-05-19T13:15:52.027Z", "content": "

Apparently I dont understand the metric because I was also thinking it should go up :D

", "rawMarkdown": "Apparently I dont understand the metric because I was also thinking it should go up :D", "votes": 1 }, { "id": 853830, "postDate": "2020-05-19T14:18:30.237Z", "content": "

Additional charts II \nScore and LB Top 15

\n\n

\"\".png?generation=1589897899410364&alt=media)

", "rawMarkdown": "Additional charts II \nScore and LB Top 15\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fe349fa28772dd401ab51b0ac75f128f2%2Fnedladdning%20(3).png?generation=1589897899410364&alt=media)\n", "votes": 1 } ] }, { "id": 851647, "postDate": "2020-05-17T19:30:34.033Z", "content": "

Update 0517

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 3 day period\n3d LB change Top 31-60 & Top 31-60 3d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev3dscore)/3))/2)PredDayAhead

\n\n

\"\"

", "rawMarkdown": "Update 0517\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F0bddc94ad756b73102769c6b31fe0bac%2F0517_1.jpg?generation=1589743688557080&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F02c5d9f6fd118a7a6522c4a7520c3809%2F0517_2.png?generation=1589743701436371&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fc5c62bf4b4bb151a381ffc2a41d937e5%2F0517_3.png?generation=1589743715975040&alt=media)\n\nAdditional scorecard -\nBest LB change Top 31-88 - 3 day period\n3d LB change Top 31-60 & Top 31-60 3d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev3dscore)/3))/2)PredDayAhead\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F659473220e376d70cb6c493f256025e7%2F0517_4.jpg?generation=1589743750645926&alt=media)\n", "votes": 1, "replies": [ { "id": 851796, "postDate": "2020-05-17T23:38:23.953Z", "content": "

Hi @kirderf, check the notebook, my model is QuantReg from statsmodels and not XGB.

", "rawMarkdown": "Hi @kirderf, check the [notebook](https://www.kaggle.com/titericz/test-1), my model is QuantReg from statsmodels and not XGB." }, { "id": 851836, "postDate": "2020-05-18T00:41:57.690Z", "content": "

@titericz perfect thanks, your notebooks I wasn't 100%, I just saw a private dataset in the beginning and imports of\nfrom sklearn.ensemble import GradientBoostingRegressor\nfrom sklearn.ensemble import GradientBoostingRegressor\nfrom lightgbm import LGBMRegressor\nimport xgboost as xgb.\nThought the dataset could have had pretrained models.\nYou will see the update tomorrow :)

", "rawMarkdown": "@titericz perfect thanks, your notebooks I wasn't 100%, I just saw a private dataset in the beginning and imports of\nfrom sklearn.ensemble import GradientBoostingRegressor\nfrom sklearn.ensemble import GradientBoostingRegressor\nfrom lightgbm import LGBMRegressor\nimport xgboost as xgb.\nThought the dataset could have had pretrained models.\nYou will see the update tomorrow :)\n\n" }, { "id": 852823, "postDate": "2020-05-18T17:47:40.523Z", "content": "

I believe my notebook needed some garbage collector. I tried many algorithms but a simple QuantReg performed best.

", "rawMarkdown": "I believe my notebook needed some garbage collector. I tried many algorithms but a simple QuantReg performed best.", "votes": 2 }, { "id": 853038, "postDate": "2020-05-18T23:16:05.477Z", "content": "

Yes it's hard to predict and follow the short trends in top/bottoms, average works better.

", "rawMarkdown": "Yes it's hard to predict and follow the short trends in top/bottoms, average works better." } ] }, { "id": 853937, "postDate": "2020-05-19T16:09:54.933Z", "content": "

Update 0519

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 5 day period\n5d LB change Top 31-60 & Top 31-60 5d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead

\n\n

\"\"

\n\n

Additional scorecard II -\n- Visualizing the score gap and Dist to the top gap\n\"\".png?generation=1589904552714636&alt=media)

\n\n

\"\".png?generation=1589904570030207&alt=media)

\n\n

\"\".png?generation=1589904682286367&alt=media)

\n\n

-Score and LB Top 15\n\"\".png?generation=1589904586617118&alt=media)

", "rawMarkdown": "Update 0519\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F118286d62b966144e6c1df1f98f810e8%2F0519_1.jpg?generation=1589904310080109&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fb3f60794d458a4b0cd80652d1799ec83%2F0519_2.png?generation=1589904323455958&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fe243a88ecab9f89db8cff60221bb56de%2F0519_3.png?generation=1589904336983765&alt=media)\n\nAdditional scorecard -\nBest LB change Top 31-88 - 5 day period\n5d LB change Top 31-60 & Top 31-60 5d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\nTodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F148b0f6981cc59a521521b2431d4c9d7%2F0519_4.jpg?generation=1589904361436285&alt=media)\n\nAdditional scorecard II -\n- Visualizing the score gap and Dist to the top gap\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F3ff263ff309877c9c214dd5a9034d144%2Fnedladdning%20(4).png?generation=1589904552714636&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F732af8dd585c850188aadf9bc58978fe%2Fnedladdning%20(5).png?generation=1589904570030207&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F33e29d5197cb66e10f30e51daa8ae6b5%2Fnedladdning%20(7).png?generation=1589904682286367&alt=media)\n\n-Score and LB Top 15\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Faf355f709fecd0527301de7ec8179573%2Fnedladdning%20(6).png?generation=1589904586617118&alt=media)\n", "votes": 2, "replies": [ { "id": 854022, "postDate": "2020-05-19T17:20:15.677Z", "content": "

Amazing how the top 4 models are in sync with each other, and almost no difference in scores between the days, good work!\nGreat work again with your model @moradnejad, this week maybe even better :)

\n\n

Look at the \"Score gap top 50\" line chart above, some models in the bottom 50 reduce their values ​​quickly, must study what algos they use.

", "rawMarkdown": "Amazing how the top 4 models are in sync with each other, and almost no difference in scores between the days, good work!\nGreat work again with your model @moradnejad, this week maybe even better :)\n\nLook at the \"Score gap top 50\" line chart above, some models in the bottom 50 reduce their values ​​quickly, must study what algos they use.", "votes": 1 }, { "id": 854100, "postDate": "2020-05-19T19:00:34.730Z", "content": "

Wow, a great day for me!

\n\n

Also, very great visualizations for tracking changes 💪 @kirderf

", "rawMarkdown": "Wow, a great day for me!\n\nAlso, very great visualizations for tracking changes 💪 @kirderf " } ] }, { "id": 850434, "postDate": "2020-05-16T16:17:07.337Z", "content": "

Update 0516

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 2 day period\n2d LB change Top 31-60 & Top 31-60 2d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\n((TodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev2dscore)/2))/2)PredDayAhead)+TodayScore(((1dAveScoreDiff+2dAveScoreDiff)/2)+1)^PredDayAhead)/2

\n\n

\"\"

", "rawMarkdown": "Update 0516\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fab3b45200263603c3c4ad2d191f2f855%2F0516_1.jpg?generation=1589645588605428&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F49b76dd42e1c6df23651ed445fb32e1d%2F0516_2.jpg?generation=1589645604774306&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F12a893513ea8d3160868519547bc7f9c%2F0516_3.png?generation=1589645625336555&alt=media)\n\nAdditional scorecard -\nBest LB change Top 31-88 - 2 day period\n2d LB change Top 31-60 & Top 31-60 2d ave. % score diff.\nPrediction Top20, 1, 3 & 5 day ahead predictions with calc\n((TodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev2dscore)/2))/2)PredDayAhead)+TodayScore(((1dAveScoreDiff+2dAveScoreDiff)/2)+1)^PredDayAhead)/2\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F754e2b51dc9ee4d7cb36a016f29b4b8b%2F0516_4.jpg?generation=1589645793785322&alt=media)", "votes": 2, "replies": [ { "id": 850478, "postDate": "2020-05-16T17:18:43.317Z", "content": "

my model seems to have some momentum :) also @bernhardklinger and @rohanrao after 2 days.

", "rawMarkdown": "my model seems to have some momentum :) also @bernhardklinger and @rohanrao after 2 days.", "votes": 2 }, { "id": 851622, "postDate": "2020-05-17T19:01:49.417Z", "content": "

Thanks for doing this again. It's fun to follow +1

", "rawMarkdown": "Thanks for doing this again. It's fun to follow +1" }, { "id": 851756, "postDate": "2020-05-17T22:15:34.770Z", "content": "

Thanks :) I think it's a good way to keep track of all movements between the models from top to bottom, identify patterns. Each team can also compare more easy and do thier own analysis based on the idea/architecture of the model.

", "rawMarkdown": "Thanks :) I think it's a good way to keep track of all movements between the models from top to bottom, identify patterns. Each team can also compare more easy and do thier own analysis based on the idea/architecture of the model.", "votes": 1 } ] }, { "id": 849432, "postDate": "2020-05-15T18:43:37.607Z", "content": "

Update 0515

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional scorecard -\nBest LB change Top 31-88 - 1 day period\n22d LB change Top 31-60 & Top 31-60 1d ave. % score diff.\nPrediction Top20 1 day ahead predictions, use actual score value change added to the new #, times days.

\n\n

\"\"

", "rawMarkdown": "Update 0515\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F8179a7818c6f91444794c2a3e5620aa5%2F0515_1.jpg?generation=1589567937497147&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fd62e50280e4710d3b0ec4eeab93a51ba%2F0515_2.jpg?generation=1589567953423085&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fc93c0bc4c271c858627868393692f296%2F0515_3.png?generation=1589568365057861&alt=media)\n\nAdditional scorecard -\nBest LB change Top 31-88 - 1 day period\n22d LB change Top 31-60 & Top 31-60 1d ave. % score diff.\nPrediction Top20 1 day ahead predictions, use actual score value change added to the new #, times days.\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F2f0524a6de944d2350d4232c016bc3e1%2F0515_4.jpg?generation=1589585947616265&alt=media)\n\n", "votes": 2, "replies": [ { "id": 849439, "postDate": "2020-05-15T18:51:02.220Z", "content": "

I've added which technique(LGB, math, Fastai etc) that are used in the models, in top 20, can be interesting to have and follow. Because a team has two submission, I just add what I see, and complements ahead to the team's total.

", "rawMarkdown": "\nI've added which technique(LGB, math, Fastai etc) that are used in the models, in top 20, can be interesting to have and follow. Because a team has two submission, I just add what I see, and complements ahead to the team's total.", "votes": 2 }, { "id": 849836, "postDate": "2020-05-16T05:33:44.263Z", "content": "

Nice that you have added which model is used for each!

\n\n

Btw you can also create Day LBs by calculating n*score[n] - (n-1)*score[n-1]

", "rawMarkdown": "Nice that you have added which model is used for each!\n\nBtw you can also create Day LBs by calculating `n*score[n] - (n-1)*score[n-1]`", "votes": 2 }, { "id": 849939, "postDate": "2020-05-16T07:34:31.553Z", "content": "

Thanks! :) I'll take that in count. I think I will try a mix of prev scorecard pred calc and with some modifications, to use in this scorecard's pred columns, for 1, 3 and 5 days pred ahead.\n((TodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)PredDayAhead)+TodayScore(((1dAveScoreDiff+5dAveScoreDiff)/2)+1)^PredDayAhead)/2

", "rawMarkdown": "Thanks! :) I'll take that in count. I think I will try a mix of prev scorecard pred calc and with some modifications, to use in this scorecard's pred columns, for 1, 3 and 5 days pred ahead.\n((TodayScore+(((TodayScore-YesterdayScore)+((TodayScore-prev5dscore)/5))/2)*PredDayAhead)+TodayScore*(((1dAveScoreDiff+5dAveScoreDiff)/2)+1)^PredDayAhead)/2" }, { "id": 850582, "postDate": "2020-05-16T19:38:15.747Z", "content": "

@kirderf Thanks for maintaining this. I haven't had a chance to look at other solutions, but I was just wondering aren't there any ensemble models being used?

", "rawMarkdown": "@kirderf Thanks for maintaining this. I haven't had a chance to look at other solutions, but I was just wondering aren't there any ensemble models being used?", "votes": 1 }, { "id": 851018, "postDate": "2020-05-17T09:23:40.147Z", "content": "

@rohanrao thanks, I'll continue as long there is an interest for it :) No not many what I can see, they with larger teams look to ensemble each other's models. Otherwise most weighted average ensembles of the same model / ML.

", "rawMarkdown": "@rohanrao thanks, I'll continue as long there is an interest for it :) No not many what I can see, they with larger teams look to ensemble each other's models. Otherwise most weighted average ensembles of the same model / ML." } ] }, { "id": 881295, "postDate": "2020-06-10T22:24:00.420Z", "content": "

Update 0610\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591827776863571&alt=media)\n\"\".png?generation=1591827775838966&alt=media)\n\"\".png?generation=1591827775822026&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591827817311542&alt=media)\nScore distribution all teams from start\n\"\".png?generation=1591827835180455&alt=media)

", "rawMarkdown": "Update 0610\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F90f4181db0303f0760564ae71676112b%2F610_1.jpg?generation=1591827478438941&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F82a6fc5e7a4e8bf4114eff867f5414ae%2F610_2.png?generation=1591827479359922&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F7e140636af35c1d5c05a7a61cefc5973%2F610_3.png?generation=1591827478155263&alt=media)\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Ff1789e7134aae1cbb9abe170a3fc8db8%2Fnedladdning%20(53).png?generation=1591827776863571&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fde8e074b4619ade8808ca0141146672a%2Fnedladdning%20(54).png?generation=1591827775838966&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F55330ebae4922f6474e331fc02ba85f5%2Fnedladdning%20(55).png?generation=1591827775822026&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F2c8d60168bbd0123b2cb102acd3e7143%2Fnedladdning%20(56).png?generation=1591827817311542&alt=media)\nScore distribution all teams from start\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F7e1fd86f3d79a43594c20cf42013b7da%2Fnedladdning%20(57).png?generation=1591827835180455&alt=media)" }, { "id": 879830, "postDate": "2020-06-09T18:48:07.927Z", "content": "

Update 0609\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591728373043015&alt=media)\n\"\".png?generation=1591728372916590&alt=media)\n\"\".png?generation=1591728375088400&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591728457203778&alt=media)\nScore distribution all teams from start\n\"\".png?generation=1591728484912030&alt=media)

", "rawMarkdown": "Update 0609\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F5a3e399db7e1ab1447934c027601894a%2F69_1.jpg?generation=1591728321567328&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F2b0174581c00c4f88f7dbf61f18c98a7%2F69_2.png?generation=1591728321325162&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F3ee3bbebc3d46c3d9c9e1a4f89f89eb3%2F69_3.png?generation=1591728321461290&alt=media)\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F336d2250a6360a92cb61466bd8915c4b%2Fnedladdning%20(48).png?generation=1591728373043015&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F5c0edf5049be5fc46954021e4c71929d%2Fnedladdning%20(49).png?generation=1591728372916590&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F3260763057608b0855a3daf8a48dfe40%2Fnedladdning%20(50).png?generation=1591728375088400&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F4d522b0751ba85ed33374442c4bfeb4d%2Fnedladdning%20(51).png?generation=1591728457203778&alt=media)\nScore distribution all teams from start\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fb21099acc2ca544a499daee80ad26c37%2Fnedladdning%20(52).png?generation=1591728484912030&alt=media)" }, { "id": 878852, "postDate": "2020-06-08T23:35:27.980Z", "content": "

Update 0608\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591659267032290&alt=media)\n\"\".png?generation=1591659265455708&alt=media)\n\"\".png?generation=1591659266694443&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591659298022027&alt=media)\nScore distribution all teams from start\n\"\".png?generation=1591659322345161&alt=media)

", "rawMarkdown": "Update 0608\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F34821042ba2e7cd6efd4542d22533aa9%2F68_1.jpg?generation=1591659224336278&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F745a16dd6d90eb7a7c8ac9a1c24867df%2F68_2.png?generation=1591659222389372&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F14feaf2fcc8a2a77a199429d3cf1d32f%2F68_3.png?generation=1591659225481371&alt=media)\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F4af6454cd4c37c0e6fbb638c7e391f01%2Fnedladdning%20(43).png?generation=1591659267032290&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fcf34da06fd548119af33c40e097f5f4e%2Fnedladdning%20(44).png?generation=1591659265455708&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F0ad7217a95cf89f2ce2672aaaf4bed85%2Fnedladdning%20(45).png?generation=1591659266694443&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fb3e360e168aa87840c7065248c4a333c%2Fnedladdning%20(46).png?generation=1591659298022027&alt=media)\nScore distribution all teams from start\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fc76038bf514ef68e052639fce29e92ff%2Fnedladdning%20(47).png?generation=1591659322345161&alt=media)" }, { "id": 877741, "postDate": "2020-06-07T23:01:00.953Z", "content": "

Update 0607\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591570712872617&alt=media)\n\"\".png?generation=1591570711698441&alt=media)\n\"\".png?generation=1591570711094114&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591570753800488&alt=media)\nScore distribution all teams from start\n\"\".png?generation=1591570856410573&alt=media)

", "rawMarkdown": "Update 0607\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fdf26c0fdb482000dd5927d4ffe96d5d4%2F67_1.jpg?generation=1591570668725292&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fd155f25dd468ef67c4489591bf91beb5%2F67_2.png?generation=1591570668631761&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F1f356d83c55a3d4ccb66907d48f6623a%2F67_3.png?generation=1591570668660355&alt=media)\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fb6b2f6bf3aab3372dfb97386bfdcd787%2Fnedladdning%20(38).png?generation=1591570712872617&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F7d2f972ae3dc022c7e5ccbeae2156ac3%2Fnedladdning%20(39).png?generation=1591570711698441&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F2debff51b0e8c6dae0b9f19da98a8446%2Fnedladdning%20(40).png?generation=1591570711094114&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fac2a4e9edafdda5b2f72bc01e8710ab8%2Fnedladdning%20(42).png?generation=1591570753800488&alt=media)\nScore distribution all teams from start\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F4a4a832de3eec4b56651ffdb48fb3d93%2Fnedladdning%20(41).png?generation=1591570856410573&alt=media)" }, { "id": 876429, "postDate": "2020-06-06T17:17:08.647Z", "content": "

Update 0606\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591464311422864&alt=media)\n\"\".png?generation=1591463760243147&alt=media)\n\"\".png?generation=1591463761005217&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591463808578363&alt=media)

", "rawMarkdown": "Update 0606\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F6c58dc4c645412d13ea6d51808c185ed%2F66_1.jpg?generation=1591463701269878&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Ff8e72f7edfe82bc44c61dc1ae178d75f%2F66_2.png?generation=1591463701762333&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F6067ea90193b5eda41cb5d012e9acb6a%2F66_3.png?generation=1591463701629022&alt=media)\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F177d8c954a3b155f42b4bcb7e8eb7051%2Fnedladdning%20(36).png?generation=1591464311422864&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F0827e0e44d3d015a9a2b3d7cdb4e29e8%2Fnedladdning%20(33).png?generation=1591463760243147&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fca689f7b29f962def3113ce32cbc3239%2Fnedladdning%20(34).png?generation=1591463761005217&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F60949a47efc13fe2b800a8fcf2b8c42b%2Fnedladdning%20(35).png?generation=1591463808578363&alt=media)" }, { "id": 875352, "postDate": "2020-06-05T17:31:53.317Z", "content": "

Update 0605

\n\n

\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591378269792155&alt=media)\n\"\".png?generation=1591378269265438&alt=media)\n\"\".png?generation=1591378271765835&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591378307733273&alt=media)

", "rawMarkdown": "Update 0605\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F6aba1ceae04d52c99c1cd7a3e57df75e%2F65_1.jpg?generation=1591377980446317&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F06cc431c861c4c2d914f86efc8e2b9fa%2F65_2.png?generation=1591377982008419&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F2e33fea0555026d5f6cf3ad835387c12%2F65_3.png?generation=1591377980157752&alt=media)\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F2f80bbe7de5adbc53ab5a6c3f966ac4d%2Fnedladdning%20(28).png?generation=1591378269792155&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F7e66b036de2c9334679593cd30d6db15%2Fnedladdning%20(29).png?generation=1591378269265438&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F2dc09c939fac727d1b55bd0ea5f85869%2Fnedladdning%20(30).png?generation=1591378271765835&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F490eb9f8dd0c448b732b66ab6386224d%2Fnedladdning%20(31).png?generation=1591378307733273&alt=media)" }, { "id": 873901, "postDate": "2020-06-04T14:02:54.493Z", "content": "

Update 0604\n\"\"\n\"\"\n\"\"\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591279348929407&alt=media)\n\"\".png?generation=1591279349624742&alt=media)\n\"\".png?generation=1591279349144642&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591279349469668&alt=media)

", "rawMarkdown": "Update 0604\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F1988c0cd57f941b457a84e3c44047d9e%2F64_1.jpg?generation=1591279303480293&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F6fb57facad69fd0f0b7043e64f90d721%2F64_2.png?generation=1591279303458333&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F340a7ce961b9f45483a0581c6b55106f%2F64_3.png?generation=1591279303204760&alt=media)\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fdd8331334acbc5fb16584083f1a0cca0%2Fnedladdning%20(68).png?generation=1591279348929407&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F80ef27c19762cd0523781b7e98f3dad5%2Fnedladdning%20(69).png?generation=1591279349624742&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F57d81957779246910927b819542d2b1f%2Fnedladdning%20(70).png?generation=1591279349144642&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F5415dda0d18772ce2cffe64774c1285e%2Fnedladdning%20(71).png?generation=1591279349469668&alt=media)" }, { "id": 872970, "postDate": "2020-06-03T16:50:46.773Z", "content": "

Update 0603\n\"\"\n\"\"\n\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591203009300088&alt=media)\n\"\".png?generation=1591203009726542&alt=media)\n\"\".png?generation=1591203009074868&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591203009488129&alt=media)

", "rawMarkdown": "Update 0603\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fb41aaef73ec804bb8197169350c16915%2F63_1.jpg?generation=1591209853544136&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F2954405c8f9b7ef3464b3b7e44ca0695%2F63_2.png?generation=1591209855794134&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F9bb9090e58f50b7788950adc844cef63%2F63_3.png?generation=1591209855862073&alt=media)\n\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fad3138a04e977fe99ad25379ef249896%2Fnedladdning%20(64).png?generation=1591203009300088&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F42bb6a4189893d5ad5b977ec1701b0ab%2Fnedladdning%20(65).png?generation=1591203009726542&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F45a97a91aeab667105b161aa9f35a135%2Fnedladdning%20(66).png?generation=1591203009074868&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F29fbececc063d0f476eba3647fcf1a89%2Fnedladdning%20(67).png?generation=1591203009488129&alt=media)", "replies": [ { "id": 873034, "postDate": "2020-06-03T17:57:34.733Z", "content": "

What happened to the first chart? :)

\n\n

Thanks for posting this regularly!

", "rawMarkdown": "What happened to the first chart? :)\n\nThanks for posting this regularly!", "votes": 1 }, { "id": 873051, "postDate": "2020-06-03T18:18:33.420Z", "content": "

It will come soon :)

", "rawMarkdown": "It will come soon :)", "votes": 1 } ] }, { "id": 871703, "postDate": "2020-06-02T15:24:34.703Z", "content": "

Update 0602\nAdditional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591111418284109&alt=media)\n\"\".png?generation=1591111418692065&alt=media)\n\"\".png?generation=1591111420200103&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591111418137060&alt=media)

", "rawMarkdown": "Update 0602\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F706b6d0ec6a5cefde8c2b98066a26b36%2Fnedladdning%20(60).png?generation=1591111418284109&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F86166600d9f07d8c1eaf143f5e078a92%2Fnedladdning%20(61).png?generation=1591111418692065&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fd8def157fddb04acd3d7cb6382e9962e%2Fnedladdning%20(62).png?generation=1591111420200103&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F649ecebb0da80d97c1b42d28bda50927%2Fnedladdning%20(63).png?generation=1591111418137060&alt=media)" }, { "id": 870519, "postDate": "2020-06-01T18:05:10.733Z", "content": "

Update 0601

\n\n

\"\"\n\"\"\n\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1591034761254106&alt=media)\n\"\".png?generation=1591034757575379&alt=media)\n\"\".png?generation=1591034759622582&alt=media)\nScore and LB Top 20\n\"\".png?generation=1591034796853624&alt=media)

", "rawMarkdown": "Update 0601\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fdee4b43ade4c2e703013c8938a00115e%2F601_1.jpg?generation=1591034610405576&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fecbe2a91b9dc71dfaec5485d668ce435%2F601_2.png?generation=1591034610006555&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F5bdff27d9935e36ef9684c2d1b75798b%2F601_3.png?generation=1591034609762501&alt=media)\n\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F03f95cf4abf4987b76993a42850dfa16%2Fnedladdning%20(56).png?generation=1591034761254106&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Ffd73fac0475ac14a12abebb7b30adba1%2Fnedladdning%20(57).png?generation=1591034757575379&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F8cd8159b59ee4656e7953a2261cb1d47%2Fnedladdning%20(58).png?generation=1591034759622582&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F5dd98caecaa12076d643105bc9bebc93%2Fnedladdning%20(59).png?generation=1591034796853624&alt=media)\n" }, { "id": 869164, "postDate": "2020-05-31T19:28:25.073Z", "content": "

Update 0531

\n\n

\"\"\n\"\"\n\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top

\n\n

\"\".png?generation=1590953275475090&alt=media)\n\"\".png?generation=1590953276248377&alt=media)\n\"\".png?generation=1590953275777142&alt=media)\nScore and LB Top 20\n\"\".png?generation=1590953275909618&alt=media)

", "rawMarkdown": "Update 0531\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F6e52957accb0f4ddaa5ec9100ceea89c%2F531_1.jpg?generation=1590953613062425&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Ffe3619a468a951380db54a2f615cecb5%2F531_2.png?generation=1590953239744354&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Ff63dd3c00369ed9f027d05a9351adcea%2F531_3.png?generation=1590953239952443&alt=media)\n\nAdditional charts -\nVisualizing the scores & Dist to the top\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F41a075ba364eed2de54e47a850424462%2Fnedladdning%20(52).png?generation=1590953275475090&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fa780a68fc24c39568afaeb79e6caa339%2Fnedladdning%20(53).png?generation=1590953276248377&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F3f12cd169eac124b422343ad8ceda5a5%2Fnedladdning%20(54).png?generation=1590953275777142&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fc1fae9364068c8a4bfffaaf15ee4b5c0%2Fnedladdning%20(55).png?generation=1590953275909618&alt=media)" }, { "id": 866819, "postDate": "2020-05-29T17:48:11.207Z", "content": "

Update 0529

\n\n

\"\"\n\"\"\n\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top\n\"\".png?generation=1590774460154868&alt=media)\n\"\".png?generation=1590774460125443&alt=media)\n\"\".png?generation=1590774460457662&alt=media)\nScore and LB Top 20\n\"\".png?generation=1590774460161138&alt=media)

", "rawMarkdown": "Update 0529\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F52a65999d593e607ab170b31783e4991%2F529_1.jpg?generation=1590774426604795&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F28e9c77c1ee2ebf19e442e6b2d91c973%2F529_3.png?generation=1590774427412137&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fad7904737e713dd759e66501d17b8241%2F529_2.png?generation=1590774427355772&alt=media)\n\nAdditional charts -\nVisualizing the scores & Dist to the top\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F1b332a933b7e5de1d852a0c6b0d6b7c4%2Fnedladdning%20(44).png?generation=1590774460154868&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fb10adf26956acd06533b8f61ebc2273a%2Fnedladdning%20(45).png?generation=1590774460125443&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fbb85739ac3accfad797dc1492b432c81%2Fnedladdning%20(46).png?generation=1590774460457662&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F883af179f4ebfd4e02570a122c401ad7%2Fnedladdning%20(47).png?generation=1590774460161138&alt=media)\n" }, { "id": 865533, "postDate": "2020-05-28T17:24:42.230Z", "content": "

Update 0528

\n\n

Halfway into w5 we can see some patterns. So far in W5 it looks like an ensemble with the top 3 models from teams DiDoKaPs/disinfectant injection/Lockdown / Northquay would be the model to use and maybe with an increasing and decreasing ensemble weight, with DiDoKaPs and also add Bernhard Klinger and Beluga as Long term models.\nI think my own model could predict with more accuracy using NN the first days. The model may also work as one of more ensemble models to balance the predicted trends with the RF average.

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top

\n\n

\"\".png?generation=1590686488627164&alt=media)\n\"\".png?generation=1590686505247899&alt=media)\n\"\".png?generation=1590686505659542&alt=media)\nScore and LB Top 20\n\"\".png?generation=1590686504696759&alt=media)

", "rawMarkdown": "Update 0528\n\nHalfway into w5 we can see some patterns. So far in W5 it looks like an ensemble with the top 3 models from teams DiDoKaPs/disinfectant injection/Lockdown / Northquay would be the model to use and maybe with an increasing and decreasing ensemble weight, with DiDoKaPs and also add Bernhard Klinger and Beluga as Long term models.\nI think my own model could predict with more accuracy using NN the first days. The model may also work as one of more ensemble models to balance the predicted trends with the RF average.\n\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Febee6c3a32f9ecaa1a95d07a641a4cab%2F528_1.jpg?generation=1590686401792662&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F887154bd680f71b9b34732a180c61e89%2F528_2.png?generation=1590686437268926&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F40563ac8295ddcbed93440bbe6cc9926%2F528_3.png?generation=1590686459062886&alt=media)\n\nAdditional charts -\nVisualizing the scores & Dist to the top\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Ffa2c2c4e0b75a0fbaa84a55c89d21d25%2Fnedladdning%20(40).png?generation=1590686488627164&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F4b3aa9675cd4b863a3b9c1fc24530627%2Fnedladdning%20(41).png?generation=1590686505247899&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F2ed3d07d272a13b8565353a8f0f2a325%2Fnedladdning%20(42).png?generation=1590686505659542&alt=media)\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F8f9c27f10915172f14d503dbb596f100%2Fnedladdning%20(43).png?generation=1590686504696759&alt=media)" }, { "id": 862539, "postDate": "2020-05-26T16:42:49.243Z", "content": "

Update 0526

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores & Dist to the top

\n\n

\"\".png?generation=1590511320541250&alt=media)

\n\n

\"\".png?generation=1590511337419125&alt=media)

\n\n

\"\".png?generation=1590511352099860&alt=media)

\n\n

Score and LB Top 20\n\"\".png?generation=1590511366180353&alt=media)

", "rawMarkdown": "Update 0526\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F06c8ae039d38aa4cffd386b14f99f7e9%2F526_1.jpg?generation=1590511148341097&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F725031f2571db2514c7d7992431689c4%2F526_2.png?generation=1590511163926317&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F55a09fd12d3df4ad8d0946220ebe58e0%2F526_3.png?generation=1590511177933053&alt=media)\n\nAdditional charts -\nVisualizing the scores & Dist to the top\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F53de3360b1e35096e72da6cd054013be%2Fnedladdning%20(32).png?generation=1590511320541250&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F21c1477eee160798f0c89c1dd8381706%2Fnedladdning%20(33).png?generation=1590511337419125&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F04b7dd2be91f68e9287791662998bfa6%2Fnedladdning%20(34).png?generation=1590511352099860&alt=media)\n\nScore and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F57db7e9182629ab82ef9c07ba1c521a9%2Fnedladdning%20(35).png?generation=1590511366180353&alt=media)\n" }, { "id": 858722, "postDate": "2020-05-23T18:22:32.020Z", "content": "

Update 0523

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the scores and Dist to the top

\n\n

\"\".png?generation=1590258089536013&alt=media)

\n\n

\"\".png?generation=1590258106365435&alt=media)

\n\n

\"\".png?generation=1590258120804038&alt=media)

\n\n

-Score and LB Top 20\n\"\".png?generation=1590258145933555&alt=media)

", "rawMarkdown": "Update 0523\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F70663415d5c9e745e1e2b01aa096747b%2F0523_1.jpg?generation=1590258031451646&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F1f4f28797f446e4dcb78533993cc62aa%2F0523_2.png?generation=1590258044456014&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fb8e78ed837705257d10350ed8b967b2a%2F0523_3.png?generation=1590258057651853&alt=media)\n\nAdditional charts -\nVisualizing the scores and Dist to the top\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F07addb5f2d6766bc281c4da937d0a392%2Fnedladdning%20(20).png?generation=1590258089536013&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F0516ca9e944d62e5cfae9c9cf9acef91%2Fnedladdning%20(21).png?generation=1590258106365435&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2Fe4ce538788f6df0068d20d80f80fa105%2Fnedladdning%20(22).png?generation=1590258120804038&alt=media)\n\n-Score and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F43e8bda7380835382a3071b3557fc637%2Fnedladdning%20(23).png?generation=1590258145933555&alt=media)\n" }, { "id": 858313, "postDate": "2020-05-23T11:19:00.393Z", "content": "

Update 0522 - reduced number of charts due to 38 teams missing in the top 50-88 in yesterday's LB

\n\n

\"\"

\n\n

\"\"

\n\n

\"\"

\n\n

Additional charts -\nVisualizing the score gap and Dist to the top gap\n\"\".png?generation=1590232636721717&alt=media)

\n\n

\"\".png?generation=1590232650764779&alt=media)

\n\n

\"\".png?generation=1590232664751648&alt=media)

\n\n

-Score and LB Top 20\n\"\".png?generation=1590232697768089&alt=media)

", "rawMarkdown": "Update 0522 - reduced number of charts due to 38 teams missing in the top 50-88 in yesterday's LB\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F4db6a653f15210560984daa3fe8c1c33%2F0522_1.jpg?generation=1590232473516496&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F478f67008cdb17bd0cdb8a8258f6eef9%2F0522_2.png?generation=1590232485337598&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F3b696cad9cfe003083aebe5d34bc18fb%2F0522_3.png?generation=1590232499086793&alt=media)\n\nAdditional charts -\nVisualizing the score gap and Dist to the top gap\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F7b7a36d5c3dd1a215a5132eedb37ece0%2Fnedladdning%20(16).png?generation=1590232636721717&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F647e47f82904e4ae044e708db6e3fef0%2Fnedladdning%20(17).png?generation=1590232650764779&alt=media)\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F7ad7e4259cb24a50b13376243f535380%2Fnedladdning%20(18).png?generation=1590232664751648&alt=media)\n\n-Score and LB Top 20\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3924325%2F6b9f19dd5f7355125891ef5633f577e8%2Fnedladdning%20(19).png?generation=1590232697768089&alt=media)" }, { "id": 857523, "postDate": "2020-05-22T17:17:08.657Z", "content": "

There are 38 teams missing in the top 50-88 in todays LB so I wait with the update until I know the status.

", "rawMarkdown": "There are 38 teams missing in the top 50-88 in todays LB so I wait with the update until I know the status." } ], "complete": true, "source": "live", "fetched_at": "2026-10-19T12:19:13Z", "competition_id": 20475, "competition_slug": "covid19-global-forecasting-week-5", "team_id": 4866933, "solution_url": "/competitions/covid19-global-forecasting-week-5/writeups/kirderf-w5-lb-scorecard-analysis-update-0611" }