Skip to content

Step 3: Data Cleaning

Now that you've mastered building models, doesn't it feel simple? But often the data you get isn't always "clean". Next, let's pick out the "impurities" from it.

🎯 Select Anomalies Like Playing a Game (Lasso)

If you trust your eyes more, try our lasso tool.

lasso

  1. Open some charts in the model view, such as the T1-T2 Relationship Chart.
  2. See those red dots far outside the red circle? They are outliers!
  3. Hold down the left mouse button and draw a circle to catch them (you can also draw a blue circle for comparative analysis). ⭕️
  4. Release! These points are "captured" and become selected.

🗑️ Batch Deletion

After selecting, click the Delete Data button in the toolbar.

  • A dialog box pops up, click Show Selected Only to list all the "bad guys" you just selected.
  • After confirming, click "Delete".

💥 Bang! The abnormal data disappears! The system will automatically generate a new model, please rename it, click the Inactive button to activate it, and finally click Fit to retrain.

Let data speak, make decisions simpler.