# cli_utils

Python API: `luxonis_ml.data.utils.cli_utils`

## Functions

### check_exists

```python
def check_exists(name: str, bucket_storage: BucketStorage):
```

Ensure that a dataset exists in the selected storage.

Parameters

 * `name` (`str`): Dataset name.
 * `bucket_storage` (`BucketStorage`): Storage backend to query.

Raises

 * `typer.Exit`: If the dataset does not exist.

### complete_dataset_name

```python
def complete_dataset_name(incomplete: str) -> list[str]:
```

### get_dataset_info

```python
def get_dataset_info(dataset: LuxonisDataset) -> tuple[set[str], list[str]]:
```

### parse_split_ratio

```python
def parse_split_ratio(value: str | None = None, train: float | None = None, val: float | None = None, test: float | None = None) -> dict[str, float | int] | None:
```

Parse a CLI split-ratio argument.

Float lists represent ratios and must sum to 1.0. Integer lists represent absolute counts.

> **Example**
> ```pycon
>>> parse_split_ratio("[0.7, 0.2, 0.1]")
{'train': 0.7, 'val': 0.2, 'test': 0.1}
>>> parse_split_ratio(train=70, val=20, test=10)
{'train': 70, 'val': 20, 'test': 10}
>>> parse_split_ratio(train=0.7, val=0.2)
{'train': 0.7, 'val': 0.2, 'test': 0.1}
>>> parse_split_ratio(train=0.7, val=0.2, test=0.1)
{'train': 0.7, 'val': 0.2, 'test': 0.1}
>>> parse_split_ratio([10, 27])
Traceback (most recent call last):
...
ValueError: Split ratio must be a list of 3 values (train, val, test).
>>> parse_split_ratio(train=0.7, val=0.2, test=0.2)
Traceback (most recent call last):
...
ValueError: Split ratios must sum to 1.0; use whole numbers for counts.
```

Parameters

 * `value` (`str | None`): A string representation of a list of 3 values (train, val, test).
 * `train` (`float | None`): Optional float or int for training split.
 * `val` (`float | None`): Optional float or int for validation split.
 * `test` (`float | None`): Optional float or int for test split.

Returns

 * `dict[str, float | int] | None`: A dictionary with keys "train", "val", and "test" mapping to their respective ratios or counts, or None if no values were provided.

Raises

 * `ValueError`: If the input format is invalid, if ratios don't sum to 1.0,
 * `or if there's a mix of counts and ratios.`

### print_info

```python
def print_info(dataset: LuxonisDataset):
```