Struct CsvSensorDataset 

pub struct CsvSensorDataset { /* private fields */ }

A dataset loaded from a CSV file.

Expected CSV schema:

col_0, col_1, ..., col_33, caption
<f32>,  <f32>, ..., <f32>, "The person is walking..."

There must be exactly T × C numeric columns followed by a caption string column. If your CSV has one row per time-step, pre-aggregate to one row per sample before loading.

Implementations

impl CsvSensorDataset

pub fn from_csv<F>(path: &Path, max_seq_len: usize, tokenize: F) -> Result<Self>

where F: Fn(&str) -> Vec<i32>,

Load all rows from a CSV file.

Arguments

• path – Path to the .csv file.
• max_seq_len – Target token sequence length.
• tokenize – A closure that converts a caption string into token IDs.

Trait Implementations

impl Dataset<SensorTextItem> for CsvSensorDataset

fn get(&self, index: usize) -> Option<SensorTextItem>

Gets the item at the given index.

fn len(&self) -> usize

Gets the number of items in the dataset.

fn is_empty(&self) -> bool

Checks if the dataset is empty.

fn iter(&self) -> DatasetIterator<'_, I>

where Self: Sized,

Returns an iterator over the dataset.

fn windows(&self, size: usize) -> WindowDataset<'_, I>

where Self: Sized,

Returns a new Dataset of all the windows of length size. The windows overlap. Is empty if the input Dataset is shorter than size.