NodeMetricSeries
One metric stream on a node with mergeable window stats. key is the metric name (e.g. 'batches', 'inference_ms', 'inference_origin_ms', 'hops'). dims are optional dimensions (e.g. vector type, model name, revision). windows are keyed by window size in seconds.
Properties
| Name | Type | Description | Notes |
|---|---|---|---|
| key | str | Metric name | |
| dims | List[str] | Optional metric dimensions | [optional] |
| windows | Dict[str, WindowSample] | Rolling-window snapshots keyed by window size in seconds | [optional] |
| last_seen | float | UTC timestamp (time.time()) of the most recent sample for this series | [optional] |
Example
from amgix_client.models.node_metric_series import NodeMetricSeries
# TODO update the JSON string below
json = "{}"
# create an instance of NodeMetricSeries from a JSON string
node_metric_series_instance = NodeMetricSeries.from_json(json)
# print the JSON string representation of the object
print(NodeMetricSeries.to_json())
# convert the object into a dict
node_metric_series_dict = node_metric_series_instance.to_dict()
# create an instance of NodeMetricSeries from a dict
node_metric_series_from_dict = NodeMetricSeries.from_dict(node_metric_series_dict)