Summary View
At the Summary View, you can view the summary information of the selected data frame.
You can view the summary of columns in the selected data frame.
You can click "Summary" tab from the top tab menus.
The header area shows the column name and its data type.
You can click the downward button to see the column menu. It shows a list of operations you can perform against this column.
The chart area shows the data distribution of this column in a bar chart. The chart type changes depending on the data type of the column.
The metrics area shows the metrics of the column. It shows different metrics depending on the data type of the column.
You can click this button to view the chart in the chart view.
You can select a column by clicking a panel. You can also choose multiple panels by Shift+Click or Control(Command)+Click.
After selecting columns, you can open the column menu and select an operation to perform the operation to multiple columns at a time.
You can search columns by clicking the "Search" icon and entering a text to the search field.
In Highlight mode, you can highlight the distribution of data that you are interested in.
Click "Highlight" button in Regular mode.
Set the condition that you want highlight in the dialog.
You can see the data that meet the condition is highlighted in Blue color in charts.
You can click "% Ratio" to see the distribution in percentage.
You can click "Close" button to exit the Highlight mode.
In Correlation mode, you can see the correlations between the target column and the other columns.
Click "Correlate" button in Regular mode.
Select a column that you want to see the correlations with other columns. You can choose either a numeric column or a logical column.
You can see the correlations between the target column and the other columns.
If your data has a lot of rows, it automatically apply the data sampling for performance. If you want to change the sample size, select the size from the "Sample" pulldown menu.
You can choose the range type for the error bars from "Range" pulldown menu. The options vary depending the data type of the target column.
Numeric target column:
- 95% Conf. Interval
- 99% Conf. Interval
- 1 Standard Deviation
- 2 Standard Deviation
- 3 Standard Deviation
- Interquartile Range (IQR)
Logical target column:
- 95% Conf. Interval
- 99% Conf. Interval
You can sort columns by the metric values. You can choose which metric for sorting from "Sort" pulldown menu. The options vary depending the data type of the target column.
Numeric target column:
- R Squared
- Correlation
- P Value
Logical target column:
- AUC
- P Value
You can create an Analytics or a Chart from the column header menu. Available Analytics and Chart types vary depending the data type of the target column and the comparing column.
Chart
- Bar
- Error Bar
- Line
- Scatter (Only if a comparing column is Numeric or Date)
Analytics
- Linear Regression
- Decision Tree
- Random Forest
- XGBoost
Statistical Tests
- T Test (Only if a comparing column has 2 unique values)
- ANOVA (Only if a comparing column has 2 unique values)
- Wilcoxon Test
- Kruskal-Wallis Test
Chart
- Bar
- Error Bar
- Line
Analytics
- Logistic Regression
- Decision Tree
- Random Forest
- XGBoost
Statistical Tests
- Chi-Square Test
You can create an Analytics with multiple predictor variables from Summary View by the following steps.
- Choose columns for predictor variables by control(command)+click or shift+click.
- Select an Analytics type from the column header menu.
Available Analytics types vary depending the data type of the target column.
- Linear Regression
- Decision Tree
- Random Forest
- XGBoost
- Logistic Regression
- Decision Tree
- Random Forest
- XGBoost
You can click "Close" button to exit the Correlation mode.
Last modified 4mo ago