cropgeeks / gridscore Goto Github PK

Hi Sebastian,

After creating a multi-valued trait, I am able to collect data multiple times.
The data exports as expected, aggregating the values correctly, either by sum, average, or last value collected.
When using the heatmap, scatterplot, or statistics viz tools at the top, however, the data values do not display properly. For instance, choosing to aggregate the results by average results in values that are several orders of magnitude higher than any of the individual values.

Things I've tried

• Adding more data points appears to blow out the scale even more i.e. the "average" of 34, 45, 67 displays orders of magnitude higher than a plot with just 34 and 45, or just 34 by itself.
• Aggregating by sum shows the same problems
• Exporting the data with a multi-value aggregate shows the correct value.
• Using the Time Series vis tool will correctly show the individual values of multi-value traits, but the average value will not display on the chart, presumably due to the y-axis limits being constrained. However, a mouse-over event in the middle of the plot will trigger a pop-up with the correct average to be displayed.

I have attached screenshots of each of the plots in question.

Best

Problem

Sometimes I need to reset a trial to remove collected observations. Normally this works as expected, and reset removes the data points. However, after collecting new data, if you export to Germinate format, Gridscore populates any multivalued traits with previously collected data and any uncollected single traits with the previously collected value.

What I've done

• Create new trial, collect data, never sync to server

  • Exports original data as expected
  • Reset trial and export -> exports a blank trial
  • Collect new data and export -> exports only new data

• Create a new trial, collect data and sync

  • Exports original data as expected
  • Reset trial, sync, and export -> Pulls in original data from server as expected
  • Reset trial, do not sync, and export -> Still pulls in original data to template (does not export blank trial)
  • Reset trial, collect some data (not all traits), do not sync, and export -> Single Traits: Overwritten with new data, Multi Traits: Includes new data and old data, Single/Mutli traits not recollected after reset are overwritten with previous values.

I realize that this behavior is probably not a bug, so my question is what would be the best way to propagate a reset of values across the trial in the server. For instance, we collect data in the same trial across multiple devices but sometimes we need to reset the trial and collect a new set of values for a different purpose. I want to have a clean slate when exporting to Germinate so it doesn't grab all historical values for a trait, but only what I just recently collected. Are there any solutions to this that don't involve cloning a blank version of the trial?

The Issues

I did another Gridscore training with some colleagues yesterday, and one thing that came up several times was how multi-valued traits are currently entered and saved. Several folks said they will score 3-5 plants for a trait per plot before moving on to the next plot. However, if they enter that data right now, they need to enter the values one at a time, click "save" to exit the plot menu, and then reopen the data collection menu again to enter the rest of the values one by one.

Also, with some of the traits, folks wanted to know if they had collected the proper amount of data points in their multivalued trait. For instance, if they always measure the height of 3 plants in the plot, it would be nice to visually see somehow that they collected 3 data points for that plot. Right now the circle stays half full as long as you have collected at least one data point, and never gets filled in all the way.

Possible Solutions

For the first issue, would it make sense to have a "next" button in the plot data menu to allow folks to quickly enter the next observation for a multi-trait? I'm not sure if the way it's currently implemented is a limitation of the underlying data structure, but I thought I'd bring it up as a possible feature to add.

Secondly, when setting up a trait, could you add an optional field to set the number of expected values for a multi-trait, and once that number of points has been entered, the circle would be completely filled in?

The Issue

When collecting data on one plot or within a guided walk, sometimes it is easy to get off track in a large trial and the accession/entry for that plot, which currently shows up in the top lefthand corner of the menu, might not be the most informative way to identify the plot.

Proposed Solution

To overcome this, we could show the Row and Column number of the plot, or the next plot that appears in the guided walk.

In the above image, we could add the row and column number directly underneath the the Accession - Rep number
Some proposed methods to do this shown below:

C - 3
Row: 1
Column: 3

or

C - 3
R1C3

or

C - 3
1 - 3
(This last way seems to be the most ambiguous though...)

The Issue

When collecting data on a trait, sometimes there are discrepancies in the recording dates. Usually the date is one day off.

Example I

Yesterday evening (Mar 29th, 6pm) I collected a data in a Gridscore trial, both single and multivalued traits. In the plot menu, the multivalued trait showed 'previous value recorded on 03/29/2023' however when I actually look at the traits, it shows 3/30/203. When I export all of the data that I recorded last night, everything exported as 3/30/2023, which is a day ahead of when they were collected.

Example II

This afternoon (Mar 30th, 3-4pm) I collected data in a different test trial. It showed the previous value was recorded on 3/29/2023 which was false because I just set everything up today. When I go look at the values though, all of them show the correct collection dates. And when I export the data, they export with the correct recording date.

It seems a little all over the board with the dates.