---
{}

---
# AutoTrain Dataset for project: bam-v2

## Dataset Description

This dataset has been automatically processed by AutoTrain for project bam-v2.

### Languages

The BCP-47 code for the dataset's language is unk.

## Dataset Structure

### Data Instances

A sample from this dataset looks as follows:

```json
[
  {
    "feat_unix": 1588744800,
    "feat_date": "2020-05-06 6:00:00",
    "id": "BTC/USD",
    "feat_open": 9013.24,
    "feat_high": 9058.04,
    "feat_low": 9006.73,
    "target": 9034.0,
    "feat_Volume BTC": 1309428.98,
    "feat_Volume USD": 144.95,
    "feat_Volume U": null
  },
  {
    "feat_unix": 1623502800,
    "feat_date": "2021-06-12 13:00:00",
    "id": "BTC/USD",
    "feat_open": 36069.87,
    "feat_high": 36103.17,
    "feat_low": 35853.97,
    "target": 36010.859375,
    "feat_Volume BTC": 83.12513066,
    "feat_Volume USD": 2993407.443,
    "feat_Volume U": null
  }
]
```

### Dataset Fields

The dataset has the following fields (also called "features"):

```json
{
  "feat_unix": "Value(dtype='int64', id=None)",
  "feat_date": "Value(dtype='string', id=None)",
  "id": "Value(dtype='string', id=None)",
  "feat_open": "Value(dtype='float64', id=None)",
  "feat_high": "Value(dtype='float64', id=None)",
  "feat_low": "Value(dtype='float64', id=None)",
  "target": "Value(dtype='float32', id=None)",
  "feat_Volume BTC": "Value(dtype='float64', id=None)",
  "feat_Volume USD": "Value(dtype='float64', id=None)",
  "feat_Volume U": "Value(dtype='float64', id=None)"
}
```

### Dataset Splits

This dataset is split into a train and validation split. The split sizes are as follow:

| Split name   | Num samples         |
| ------------ | ------------------- |
| train        | 38600 |
| valid        | 9660 |