# Datascience_4_competition — Data

来源：https://www.kaggle.com/competitions/datascience-4-competition/data

采集时间：2026-10-20T08:23:07Z

## 📂 Files

| File                       | Rows × Cols | What it contains                                                        |
| -------------------------- | ----------- | ----------------------------------------------------------------------- |
| **train.csv**              | 564 × 66    | 64 binary clinical features, an **ID** column, and the target **label** |
| **test.csv**               | 143 × 65    | Same 64 features + ID (no label)                                        |
| **sample\_submission.csv** | 143× 2     | Template you will upload to Kaggle (`ID`, `label`)                      |

---

## 📑 Column Definitions

| Column                   | In files         | Type       | Description                                                                                                                         |
| ------------------------ | ---------------- | ---------- | ----------------------------------------------------------------------------------------------------------------------------------- |
| **ID** (`ID`)    | all              | int        | Unique patient identifier                                                                                                           |
| **feature0 … feature63** | all              | int (0/1)  | Binary flags for anonymised clinical indicators (symptoms, lab thresholds, codes). A value of **1** means the indicator is present. |
| **label**                | *train.csv* only | int (0–10) | Disease category to predict                                                                                                         |

No missing values are present; every feature is either **0** or **1**.

---

## 🧾 Prediction Target

Your model must predict **`label`** for every row in **test.csv**.

---

## 📥 Submission Format

Upload a CSV named however you like but with **exactly two columns**:

```
ID,label
```

* `ID` – copy of the **ID** from *test.csv*
* `label` – integer prediction 0 … 10

A valid first few lines look like:

```
ID,label
1,7
3,0
4,6
...
```

Good luck, and happy modeling!