mindspore_gl.dataset.MetrLa

class mindspore_gl.dataset.MetrLa(root)[source]

METR-LA Dataset, a source dataset for reading and parsing METR-LA dataset.

About METR-LA dataset:

METR-LA is a large-scale dataset collected from 1500 traffic loop detectors in Los Angeles country road network. This dataset includes speed, volume and occupancy data, covering approximately 3,420 miles.

Statistics:

  • Time step: 12,6850

  • Nodes: 207

  • Edges: 1515

Dataset can be download here: METR-LA .

You can organize the dataset files into the following directory structure and read by mindspore_gl.dataset.MetrLa.get_data API.

.
├── adj_mat.npy
└── node_values.npy
Parameters

root (str) – path to the root directory that contains METR-LA/adj_mat.npy and METR-LA/node_values.npy.

Inputs:

  • in_timestep (int) - numbers of input time sequence.

  • out_timestep (int) - numbers of output time sequence.

Raises

  • TypeError – if root is not a str.

  • RuntimeError – if root does not contain data files.

  • TypeError – If in_timestep or out_timestep is not a positive int.

Examples

>>> from mindspore_gl.dataset.ppi import MetrLa
>>> root = "path/to/metrla"
>>> dataset = MetrLa(root)
>>> features, labels = dataset.get_data(in_timestep, out_timestep)
get_data(in_timestep, out_timestep)[source]

Get sequence time feature and label.

Parameters

  • in_timestep (int) – numbers of input time sequence.

  • out_timestep (int) – numbers of output time sequence.

property node_count

Number of nodes.

Returns

  • int, number of node.

Examples

>>> #dataset is an instance object of Dataset
>>> node_count = dataset.node_count