Results Manager

Figure 1. Results Manager

Results Manager common terms

In Results Manager there are a number of key terms. These are mostly related to naming conventions for results, groups of results, and how those results are manipulated. See Table 1 below.

Table 1. Commonly used terms

Central workspace

The central workspace in the Results Manager window is where you will see your charts, tables, statistics and other results. For some displays, there are right click contextual menus within the central workspace. The display in the central workspace is specific to the item you have selected in the tree menu or custom chart and statistics pane - if no item has been selected, no chart will be visible.

Tree menu

The tree menu on the left side of the Results Manager window (Figure 2) shows all result sets from the current session, any saved results from previous runs, along with any loaded result files (*.res.csv) and time series files (Data file formats).

Once run, results from a scenario appear grouped together under Latest run and as a separate item grouped by scenario in run order. Those previous runs are named using the following convention, where # represents the run number from the current session:

• (#) Scenario name – eg. (2) Stage A, means that it is the second run of the current session, and that the scenario run was called Stage A. 
• (#) Scenario Input Set name – when there is only one scenario, and more than one Scenario Input set.
• (#) Scenario name (Scenario Input Set name) – when there is more than one scenario and the chosen scenario has more than one scenario input set.

The Results Manager tree menu is shown in Figure 2.

Figure 2. Results Manager, tree menu

The following should also be noted. 

• The icon displayed after a scenario run corresponds to the notification level in the Log Reporter. The icon indicates that the run contains warnings and theicon indicates that the run was incomplete because of a fatal error.
  
• Previously saved result sets and time series can be loaded (see Loading Results), and appear under Loaded Result Files and Loaded Time Series, respectively.
• You can rename loaded result sets, time series and runs (except the latest run) that appear in the tree menu. However, this does not change the name of results grouped under the result set, time series or run - they will retain the original name, and any view tab will list the full name of the result as it was loaded in the Source project. 
• Run results can be saved with the project using the Save Run in Project contextual menu item (in upper left contextual menu, Figure 2). This saves the run results and loads them automatically when the project is next opened. Note that this will increase the project's file size, particularly if you record many results or the model has a long run time. Runs that will be saved or have previously been saved in the project are indicated by a save icon next to the scenario name (Figure 3). If you have previously chosen to save a run in the project and you no longer wish to do so, you need to delete the run (right click on the run and choose Delete from the contextual menu). 

Figure 3. Tree Menu showing a run saved in the project

Result tabs

The central workspace of Results Manager contains result tabs, which can display:

• A single time series result (as in Figure 1);
• A custom chart, which allows the comparison of multiple time series; 
• A unique view specific to a non-time series result (eg. Scenario Mass Balance, Travel Time, Borrow and Payback); and
• A unique view of statistics generated for either an entire scenario (eg. Mass Balance) or for a single result from different scenarios (eg. Annual Descriptive Statistics).

Apart from results with unique views, each result tab contains the following view tabs:

• The Chart tab provides a graphical representation of data, which can be customised using the Chart tab toolbar (refer to Charts);
• The Table tab displays the results in a table and allows you to query the table using advanced search functionality (refer to Table); 
• The Transforms tab provides a list of all transforms that have been applied to the result(s) in that result tab (refer to Transforms);
• For custom charts only, the Data tab displays what results the custom chart contains in a tree menu; and
• The Statistics tab shows a statistical analysis of the data – Univariate statistics are displayed for a single result, Bivariate and Univariate comparison statistics are displayed for custom charts (refer to Statistics).

To use result tabs:

• Shift-click on a result (in the tree menu), or a custom chart or statistic (in the pane below the tree menu) to open that item in a new result tab.
• Click and drag a result tab to move it to a new result window, which will include the central workspace and view tabs depending on the type of view (see unique views above). You can repeat this to open more result windows.
• Once you've done this, and have one (or more) result window open along with the Result Manager window (which has a tree menu), you can Shift+Click on a result in the chart tab of any of these windows and drag it to another window to copy the result to the other window. This is synonymous to clicking the Drag button on the Chart toolbar and then dragging the result to another window. The result window or tab to which you dragged the result will automatically change to a custom chart, if it is not one already. 

Custom charts and statistics pane

The custom charts and statistics pane is located to the left of the central workspace, below the tree menu panel. It lists the custom charts and user-generated statistics that have been created in the Source project. The pane and its contextual menus are shown in Figure 4. The icon beside an item in the pane indicates that a transform has been applied. Note that Add Transform is only shown if there are available transforms.

Figure 4. Results Manager, Custom chart and statistics pane with defined transforms available

Toolbar

The Results Manager toolbar (Figure 5) is located near the top of the Results Manager window and includes load, save, search and transform options.

You can load a:

• folder of result files (Ctrl+R);
• time series (Ctrl+O);
• Source Database (Ctrl+D); or
• configuration file (Ctrl+L).

You can also save the current configuration (Ctrl+S) and set up your data transforms (Ctrl+Shift+T). To go to a result in the tree menu, select the search icon and then start typing a search string for results from the tree menu for the currently active scenario to appear as you type, see Searching for results. 

Figure 5. Results Manager toolbar

Custom charts

Custom charts allow you to examine and statistically analyse multiple results within and between runs, compare results to external data sources and automatically update with the latest results each time a model is run. Whenever you want to compare two or more results, you need to create a custom chart.

Creating custom charts

A custom chart comparing Results A and B can be created using one of two methods:

1. Right click on Result A and choose Add to Custom chart » New Custom chart from the contextual menu. This will create a new custom chart in the panel below (named Custom Chart #2 for example); and
2. Right click Result B and choose Add to Custom chart » Custom chart #2.

OR

1. Click on the result in the tree menu to display Result A in the chart tab; and
2. Then, Ctrl-click on Result B.

Additional results can be added to a custom chart by repeating the second step of either of these methods. There is no limit to the number of results that can be added to a custom chart, but only time series data can be added to custom charts. Note that if a custom chart is active when you Ctrl-click on a result, the result will be added to that custom chart. Otherwise, a new custom chart will be created. Ctrl-clicking on a result that is already in a custom chart will delete it.

Custom chart features

The view tabs in a custom chart differ from those of a single result as summarised below.

• Chart tab - each result is displayed in the chart and listed in the legend next to its colour, along with a checkbox, which can be toggled off to hide the result. When a custom chart is created, additional chart types become available for comparing results to each other, e.g. difference, double mass, and scatter plots. Refer to Charts and Chart Types.
• Table tab - each result is displayed in its own column relative to a date column that covers the date range for all of the data combined.
• Transforms tab - if you apply a transform to a custom chart it is applied to all results in the custom chart. Refer to data manipulation.
• Data tab - this tab only exists in a custom chart. It displays the results that have been added to the custom chart in the tree menu in the panel on the left of the Results Manager window.
• Statistics tab - in addition to univariate statistics, bivariate statistics are displayed. By default the statistics are calculated for the overlapping date range between all results in the custom chart. Refer to Statistics.

Custom charts are automatically saved when you save your project. This will save which results are in your custom chart along with the transforms you have applied, the chart type, and any other options you have configured (eg. short header in Tables). However, results themselves are not saved unless you choose to save the run(s) that generated the results. When you next open your project, the custom chart will be empty until a result is generated that exactly matches the name of a result used in the custom chart, ie. the scenario name and run number have to match, so you need to run the scenarios in the same order. If results from Latest Run are used to generate a custom chart, it will be populated as soon as you run any scenario. Therefore, if you want to return to an analysis in a custom chart after closing a project, it is best to use either results from Latest Run or to save the run in the project.

Managing Results

Import and Export From File

Results can be imported and exported from file, as discussed in Import and Export from File.

Searching for results

You can search for specific results without navigating through the tree. Searches can be undertaken in one of two ways:

• Enter the result name in the Search field above the tree menu to filter the tree; or
• Click on the Go to a result in the tree (magnifying glass) icon in the toolbar (synonymous to entering Ctrl+T).

Each of these methods searches the results differently:

• The Search field searches all levels of the tree menu (including loaded results or time series) for the entered phrase and hides any result that does not contain that phrase anywhere in its path. For example searching for 'Inflow' will leave unhidden all results with the word Inflow in their path, which includes all results with inflow in their name (eg. Total Inflow Volume), and also all results for all Inflow nodes. Therefore, this search field is most useful when searching for a specific result (eg. Downstream Flow Volume) or node name (eg. Fish Creek Confluence).
• The Go To option allows you to search the currently active scenario using the wild card * to represent any character. If there is no active scenario (eg. because you have selected a custom chart), it will search in the Latest Run. For example, "Conf*Inf" can be used to find the Total Inflow Volume for all Confluence nodes, regardless of their node names. You can then select one of the returned results and it will be displayed. This dialog displays a maximum of 16 results, even if there are more results that meet the search criteria.

Date Range and Aggregator

The Date Range and Aggregator toolbar (Figure 7) provides a quick method for selecting and displaying focused results. The Date Range filter can be applied to isolate results that either occur between the selected start and end dates, and the Repeating Range Filter can be used to select and display results that occur during a specified repeating date range, E.g. by setting the Start Date to 1 Jan and the End Date to 31 Jan all results for all January's during the model run will be displayed.

The Aggregator filter can be used to display the Mean, Maximum, Minimum or Sum of the data on an Annual, Monthly or Daily timescale. The Aggregator filter is linked to the Date Range filter and will only perform the selected aggregation method on the data that falls between the Start and End dates.

Figure 7. Results Manager, Date Range and Aggregator

Configurations

A Results Manager configuration file (*.dora) contains a record of:

• All custom charts;
• Transforms and where they are used;
• User-generated statistics;
• File locations of loaded time series and results files;
• Water Year start date (set in Project Options);
• Short table headers (set on Table view tab); and
• Visible decimal places (set in Application settings).

You can: 

• Save the current configuration of the Results Manager by selecting Save Configuration from the File menu or toolbar; and
• Load a previously saved configuration using the Load Configuration from the File menu or toolbar. The integrity of a configuration file depends on the run order being consistent.

Hot Keys

Choose Help » HotKeys to view a list of shortcuts that can be used within the Results Manager (Figure 8).

Figure 8. Results Manager, hotkeys

Stand-alone mode

Results Manager can be run independently of Source, allowing you to analyse previously saved Source results or time series from other programs without opening Source. You can launch Results Manger in stand-alone mode from your Windows Start menu, the standard location is:

Start » All Apps » eWater Source n.n.n.n » Results Manager n.n.n.n

The only difference between stand-alone mode and running Results Manager within Source is a Configuration Options button in the Results Manager toolbar (Figure 9). This opens a Configuration dialog (Figure 8) that allows you to configure the following options:

• Enable Rounding – Determines whether the values in the Table and Statistics views are rounded to the number of places entered in Visible Decimal Places.
• Short Table Headers – Toggles on or off the display of long or short names of results in the Table view. Can also be enabled on the Table view tab.
• Water Year Start – Allows you to set the default water year start date, which affects the calculation of statistics in the Statistics view.

Figure 9. Results Manager toolbar, stand-alone mode