# Find your subject — Data

来源：https://www.kaggle.com/competitions/find-your-subject/data

本地 Meta Kaggle 快照；不是本次实时页面。

# Dataset Description



## Files

*   **mlrg_features.csv** - complete dataset. It contains 20000 samples out of which first 10000 has labels mentioned in mlrg_target.csv file. These 10000 samples should be used for training and rest 10000 should be used for testing.
*   **sample_submission.csv** - a sample submission file in the correct format


## Columns
for mlrg_features.csv
*   `ID` - definition of example_id
*   `Question` - definition of feature

for mlrg_target.csv
* `ID` -  definition of example_id
* `Label` - Label of the sample

for output.csv
refer sample submission csv file

In output.csv younare required to submit predictions for all 20000 samples.