# Datathon — Data

来源：https://www.kaggle.com/competitions/datathon202/data

本地 Meta Kaggle 快照；不是本次实时页面。

# Dataset Description

The train.csv file has 17 columns , it is the training dataset , the dataset describes the presence of harmful chemical substance in the surroundings of some selected cities of USA . the target value is the "SI" column which stands for "sustainability index". the higher its value the less habitable is that city for a healthy living.
You need to predict the future value of "SI" using the other attributes.
## Files

*   **train.csv** - the training set ,it has 16000 rows , train your model on this datset.
*   **test.csv** - the test set ,it has 4000 rows , predict the target value for this dataset.
*   **sample_submission.csv** - a sample submission file in the correct format, your submission file should have the same column names as this file.


## Columns

*   `Id` - It is the series number of rows
*   `City` - City names 
*   `Date` - Date of the values recorded
*   `WP1` - It stands for Weather_parameter-1  .
*   `WP2` - It stands for Weather_parameter-2
*    `N1 ,N2,N3,N4` - These are the compounds of Nitrogen like No2,No,etc.
*    `Carbon` - level of Carbon contaminant
*    `Sulphur` - level of Sulphur contaminant
*    `Ozone` - level of Ozone 
*    `IN1,IN2` - level of Inert gases.
*    `OC1,OC2` - level of Organic compounds .
*     `SI` - Sustainability Index the higher its value the higher is the damage to health.