Benchmark
Benchmark
A time series forecasting benchmark consisting of multiple Task
objects.
Attributes:
Name | Type | Description |
---|---|---|
tasks |
list[Task]
|
Collection of tasks in the benchmark. |
Source code in src/fev/benchmark.py
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 |
|
Attributes
tasks: list[Task] = tasks
instance-attribute
Functions
from_yaml(file_path: str | Path) -> Benchmark
classmethod
Load benchmark definition from a YAML file.
The YAML file should contain the key 'tasks'
with a list of dictionaries with task definitions.
tasks:
- dataset_path: autogluon/chronos_datasets
dataset_config: m4_hourly
horizon: 24
num_windows: 2
- dataset_path: autogluon/chronos_datasets
dataset_config: monash_cif_2016
horizon: 12
Parameters:
Name | Type | Description | Default |
---|---|---|---|
file_path
|
str | Path
|
URL or path of a YAML file containing the task definitions. |
required |
Source code in src/fev/benchmark.py
24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 |
|
from_list(task_configs: list[dict]) -> Benchmark
classmethod
Load benchmark definition from a list of dictionaries.
Each dictionary must follow the schema compatible with a fev.task.Task
.
Source code in src/fev/benchmark.py
57 58 59 60 61 62 63 |
|