Filter on Date Range in Use Chart

[saalmu...@gmail.com]

Hi Ale, 

i'm trying to create a Cumulative distance per year chart where the individual data series should be the cumsum(DISTANCE) for individual years.

I found this example from Marc in an old thread mean(TSS["2015/01/01":"2015/12/31]) but it causes syntax error. 

This 

daterange("2020/01/01", "2020/05/01", metrics(DISTANCE));

doesn't work either.

How can i filter a date range in a user chart? 

 

[Ale Martinez]

Looking for the metrics() function documentation in the wiki (?) 

[saalmu...@gmail.com]

Ah!     metrics(BikeStress, "2020/07/01", "2020/07/31")

Sorry for bothering you about it, Ale.

Thank you for your fast response!

[Ale Martinez]

El jueves, 18 de febrero de 2021 a la(s) 20:56:03 UTC-3, saalmu...@gmail.com escribió:

Ah!     metrics(BikeStress, "2020/07/01", "2020/07/31")

Sorry for bothering you about it, Ale.

Thank you for your fast response!

Honestly this is getting increasingly annoying for me. I tried several strategies s.t. to remind about forum guidelines or even to use some irony, but nothing seems to work.

Since, if I keep responding the cost to ask in the forum will continue to be lower than to RTFM, I will opt for not to answer anymore to users insisting on that behavior.

[saalmu...@gmail.com]

Well, i did search but overlooked the right spot - clearly my mistake.

[saalmu...@gmail.com]

My last question - for a long while, promise!

For a selected year in Trends I want to map a metrics vector (Date) with length being count(activities) to a new Date vector that has a data point for every day in the year (length = 365).

With seq i get that vector, but mapping the actual metrics to it is the problem.

I searched the forum properly plus the net. I get halfway there using the match() function to give me a vector of indices but am stuck there.  Things like complete(), CJ, joins  and other functions don't seem work.

The ultimate purpose is to create a metrics vector for Distance that has the actual distance for the activities and the distance of the previous not empty point for those dates where no activity exists. That last part i should be able to do with sapply.

Is there a way to do this within a User Chart?

Thanks, Ale. 

[Ale Martinez]

El viernes, 19 de febrero de 2021 a la(s) 03:53:13 UTC-3, saalmu...@gmail.com escribió:

My last question - for a long while, promise!

For a selected year in Trends I want to map a metrics vector (Date) with length being count(activities) to a new Date vector that has a data point for every day in the year (length = 365).

With seq i get that vector, but mapping the actual metrics to it is the problem.

I searched the forum properly plus the net. I get halfway there using the match() function to give me a vector of indices but am stuck there.  Things like complete(), CJ, joins  and other functions don't seem work.

The ultimate purpose is to create a metrics vector for Distance that has the actual distance for the activities and the distance of the previous not empty point for those dates where no activity exists. That last part i should be able to do with sapply.

Is there a way to do this within a User Chart?

Yes

[saalmu...@gmail.com]

Ok, don't tell me - let me try and figure it out ....

[Ale Martinez]

El viernes, 19 de febrero de 2021 a la(s) 19:47:08 UTC-3, saalmu...@gmail.com escribió:

Ok, don't tell me - let me try and figure it out ....

Great, post your solution here. Mine is included in the Energy Balance Chart.

[saalmu...@gmail.com]

Hi Ale, 

- i don't think my version handles multiple activities on a single day

- the time shifting to align the curves is very inelegant 

- probably some other issues

but it looks like the graph i wanted.

I'll have a look how you did it tomorrow ....

Cheers

J

[saalmu...@gmail.com]

  I get an error when trying to post the gchart file. So here's the code for the 2019 series:

{

    init {

        xx<-c();

        yy<-c();

        FullYear<-seq(43464,43464+364,1);

        activities<-0;

}

    activity {

        # as we iterate over activities

        activities <- activities + 1;

    }

    finalise {

        # we just fetch metrics at the end

        ActivityDates       <- metrics(date,    "2019/01/01","2019/12/31");

        ActivityDistances   <- metrics(Distance,"2019/01/01","2019/12/31");

        # make a 365 long vector out of ActivityDistances

            # append distances and Dates

                Dates           <- c(FullYear+0.2,ActivityDates-0.2);

                AnnualDistance  <- c(rep(0,365),ActivityDistances);

            # sort and unique Distances and Dates by Dates

                 multisort(ascend, Dates, AnnualDistance);

                 Dates<- round(Dates);

                 multiuniq(Dates, AnnualDistance);

yy <- cumsum(AnnualDistance);

    }

    x { Dates+365; }

    y { yy; }

}

[saalmu...@gmail.com]

Hey Ale, 
as promised you haven't heard from me in a long time. I hope this finds you well?

I'm curios to know for what release and at what point in time the "Plan" function can be expected to be released?

Cheers

Juergen

[Mark Liversedge]

3.7.

There is a project here:

https://github.com/GoldenCheetah/GoldenCheetah/projects/2

The e2e process is described here, with links to the deck:

http://markliversedge.blogspot.com/2020/02/an-athlete-performance-management.html

I will likely to start with modelling adaptation before moving on to interventions and then the ui/ux for it (diary, automation etc)

Mark

[saalmu...@gmail.com]

Thanks, Mark !!

[saalmu...@gmail.com]

Hi Ale, 

I hope you're well. 

Recently transferred all GC from Win10 to Linux Mint. All athlete data, charts and layouts came across well and work, which is great.

The only thing i can't find is where the filters are stored. I have quite a number as well as "set" actions in there. How do i get those across to linux?

Also, is it possible to use an ERG file as a 'segment' or a power test so i can compare performance on the same ERG file over time?

Happy to open separate threads , let me know.

Cheers

Juergen