# Forest Fire Prediction — Data

来源：https://www.kaggle.com/competitions/forest-fire-prediction/data

采集时间：2026-10-20T11:09:16Z

The data divides for training and testing are named train.csv and test.csv, respectively. There are **200** instances in the training split and **50** examples in the testing split. The complete descriptions for these splits are given below.

## train.csv (200 examples)
The training set consists of the following feature columns:

1. Day (Data type - INT)
2. Month (Data type - INT)
3. Year (Data type - INT)
4. Temperature (Data type - INT)
5. RH (Data type - INT)
6. Ws (Data type - INT)
7.  Rain (Data type - FLOAT)
8. FFMC (Data type - FLOAT)
9. DMC (Data type - FLOAT)
10. DC (Data type - FLOAT)
11. ISI (Data type - FLOAT)
12. BUI (Data type - FLOAT)
13. FWI (Data type - FLOAT)


The column - "**Classes (Data type - string)**" is the label column. 

## test.csv (50 examples)
The test split includes all of the training dataset's feature columns. Your task is to forecast the labels (**Classes** column) for each of these cases corresponding to their respective ID and submit the data in the same style as **sample_submission.csv**.

**Important** - Please keep in mind that the numbers in sample_submission.csv are only placeholders; you must build your own submission file for the whole test set.