Coronavirus Dashboard With Automated pdf

[MauroCapolino]

Hello I created based on your data and the Italian data a dashboard that I update daily with over 50 analysis. I also create automatically a pdf with all the reports. Happy to help you in case you want to use part of it
The dashboard is on google data studio

https://datastudio.google.com/u/0/reporting/9ce371f1-62fa-4f2e-8d3c-da4d6dff967b

From here there is also a automated report in pdf that I sent out every day if you like to be copied let me know

[JiPiBi]

Seems great but, I get this error when linking :

Erreur de configuration de l'ensemble de données

Impossible d'accéder à votre ensemble de données depuis Data Studio.

The only issue could be the global reliability of data here (for Italian values I have no doubt)

[MauroCapolino]

The global data is under update since I had to change my dB this morning to adapt to tew John Hopkins format Will be online in 1 hr Let me know email if you like to get pdf Mauro

…

On 22 Mar 2020, at 12:20, JiPiBi ***@***.***> wrote:  Seems great but, I get this error when linking : Erreur de configuration de l'ensemble de données Impossible d'accéder à votre ensemble de données depuis Data Studio. The only issue could be the global reliability of data here (for Italian values I have no doubt) — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[joaopedrosantos]

interesting, will look at google data studio. any tips to learn?

[JiPiBi]

@MauroCapolino
Now I have access
But I dont understand your calculations for each country
I understand that you give the daily value for the 2 last days , but on which basis are the percentages calculated : from one day vs the previous ?

And mainly what represent the total (as the values are cumulative why have a total ??)
and for the percentage on the line total general , on what is it based .

For Netherlands 363900% and United Kingdom 0% are quite strane too ??

The same for the deaths ....

Edit : at first look I did nt see all the different views , very impressive
if you could explain how all that is managed or give a link that explains this tool

[MauroCapolino]

Yes incremental is vs previous day Yesterday I did major review and fix Check again Mauro

…

On 22 Mar 2020, at 22:25, JiPiBi ***@***.***> wrote:  @MauroCapolino Now I have access But I dont understand your calculations for each country I understand that you give the daily value for the 2 last days , but on which basis are the percentages calculated : from one day vs the previous ? And mainly what represent the total (as the values are cumulative why have a total ??) and for the percentage on the line total general , on what is it based . For Netherlands 363900% and United Kingdom 0% are quite strane too ?? The same for the deaths .... — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[MauroCapolino]

Just open an account put some data in a google sheets and then from studi link it to the sheets the rest is basic reporting Mauro

…

On 22 Mar 2020, at 20:00, joaopedrosantos ***@***.***> wrote:  interesting, will look at google data studio. any tips to learn? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[JiPiBi]

In first tab , you are adding all the time series values for each country which en fact are cumulative and not daily values : so it needs to be fixed

[MauroCapolino]

Please send me the page name so that I can check I have 57 pages Mauro

…

On 23 Mar 2020, at 18:57, JiPiBi ***@***.***> wrote:  In first tab , you are adding all the time series values for each country which en fact are cumulative and not daily values : so it needs to be fixed — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[JiPiBi]

As I said it is the first , look for example for France on the third value in first col , I repeat that you add cumulative values from the beginning , instead of keeping the last one , I summed up to confirm your value.

[MauroCapolino]

I do not know what is the issue Nella mia pagina iniziale i casi totali della Francia sono 20123, Italy 63927... Che e’ lo stesso dato del sito Johm Hopkins totale casi Non vedo il problema! On 24 Mar 2020, at 06:15, JiPiBi <notifications@github.com> wrote: As I said it is the first , look for example for France on the third value in first col , I repeat that you add cumulative values from the beginning , instead of keeping the last one , I summed up to confirm your value. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1223 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AOZ2JKYM6LUY4WMLJOSK7YTRJA6XZANCNFSM4LRHIADQ>.

[JiPiBi]

La mia famiglia e venuta nella Francia da San Giovanni Bianco
Parlo un po l'italiano, ma non troppo e con molta falta :-), as my english in fact :-(

Your dashboard has changed now , so it seems you fixed the issue

In my last check, for France the third line was labelled Total gen and was the sum of all cumulative value .

I will go on explore your numerous pages
Is there a mean to send you my remarks not in this github repo (directly on your site ?)

[MauroCapolino]

Yes I did major fix work and now there is a mobile version at the link in the orange part try it On 24 Mar 2020, at 11:49, JiPiBi <notifications@github.com> wrote: La mia famiglia e venuta nella Francia da San Giovanni Bianco Parlo un po l'italiano, ma non troppo e con molta falta :-), as my english in fact :-( Your dashboard has changed now , so it seems you fixed the issue <https://user-images.githubusercontent.com/44967630/77416691-7c9e7b00-6dc4-11ea-8e71-68f41643421b.png> In my last check, for France the third line was labelled Total gen and was the sum of all cumulative value . I will go on explore your numerous pages Is there a mean to send you my remarks not in this github repo (directly on your site ?) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1223 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AOZ2JK6V6HXZBXFOOK77V5LRJCF2TANCNFSM4LRHIADQ>.

[JiPiBi]

In page 4 , what disturbs me is that I think that France and Italy are following the same curve (the slopes seem the same)
but I can't really compare them because the beginning of epidemics is in hard code

At the beginning the values are small , and the true date of the beginning is not easy to choose, you could also say that the first cases in France were earlier , why not ( in fact it was end of january)

So can I modify myself this beginning date of a country to compare the trends ?

[JiPiBi]

A suggestion : Is it possible to go directly to a given page, because there are a lot and you have your favourite ones (by a dropdown list ?)

[MauroCapolino]

Changed the menu to show all the pages on the left( you can close the list of page with the arrow on top right of the list) Let me know if you like it Mauro

…

On 24 Mar 2020, at 13:56, JiPiBi ***@***.***> wrote:  A suggestion : Is it possible to go directly to a given page, because there are a lot and you have your favourite ones (by a dropdown list ?) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[MauroCapolino]

The model is one so if I allow you to change it would change for all the users Sorry the tool does not allow per user parameters input. If I add a linked field to the model with the start date and you change it than all can change it causing conflict I anyhow do not consider relevant the first 10 relative days since all the number are low what is important is how each country accelerates like USA Mauro

…

On 24 Mar 2020, at 13:06, JiPiBi ***@***.***> wrote:  In page 4 , what disturbs me is that I think that France and Italy are following the same curve (the slopes seem the same) but I can't really compare them because the beginning of epidemics is in hard code At the beginning the values are small , and the true date of the beginning is not easy to choose, you could also say that the first cases in France were earlier , why not ( in fact it was end of january) So can I modify myself this beginning date of a country to compare the trends ? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[JiPiBi]

Sorry for my late answer , I was struggling with new detailled data for France that give at a departement level hospitalized and ICU (data you already have for Italy at a region level )

For me now what is important is to know hosp + ICU for all the countries : I wrote an issue on that #1677
I hope you share my point of view :-)

About your dashboard , the improvment with left menu is great !

I'm trying now to learn DataStudio to publish data , because python is great for anlysis and cleaning of data , but views are poorly illustrated

[JiPiBi]

Your charts are based on a given date that is supposed to be the start of the epidemic in a country.
Dont you think that it will be interesting to measure the confinment impact ?

Another question :
For me the new hospitalized and the new ICU for a given day is an important data
In France as data are given mainly for hospitals I think that I can measure the new hospitalized people , more difficult for ICU because people in ICU can deceased or stay in hospital , so the movments or flows are not clear to me ( I can suppose that deceased came from ICU but with the difficult choices they have to make , not completely sure )
What do you think of that KPI ?