Icon

test1

1. Load Data

CSV Reader imports the climate dataset and should auto-detect delimiter, encoding, header, and data types.

2. Data Exploration

Statistics View inspects structure and distributions. Statistics generates summary metrics. GroupBy compares values by Year and Country.

3. Data Preprocessing

Missing Value handles nulls, Duplicate Row Filter removes repeated records, String to Number fixes numeric imports, One to Many encodes Country, and Column Filter keeps model features. A categorical target should be created here if needed.

4-6. ML Training, Application, Evaluation

Partition data into 80/20, train a Decision Tree on the training split, predict on validation data, and evaluate with Scorer using confusion matrix and accuracy.

Load climate CSV dataset
CSV Reader
Data Explorer view of loaded data
Statistics View
Row Filter
Descriptive statistics for all columns
Statistics
Compare climate measures by Year and Country
GroupBy
Handle missing values
Missing Value
Remove duplicate rows if present
Duplicate Row Filter
Convert numeric-looking string columns to numbers if needed
String to Number
Bar Chart
Split data into 80 percent training and 20 percent validation
Table Partitioner
Numeric Scorer
One hot encode Country if needed
One to Many
Evaluate classification accuracy and confusion matrix
Scorer
Keep relevant predictor variables
Column Filter
Sorter
Train decision tree on training partition
Decision Tree Learner
Apply decision tree model to validation partition
Decision Tree Predictor
Line Plot
Rule Engine

Nodes

Extensions

  • No modules found

Links