Questions tagged [metrics]
Software metrics are quantitative data related to software.
metrics
2,880
questions
0
votes
0
answers
19
views
why does the f1 score declines from 1.0 to 0.0?
In the evaluation of a multi-class YOLOv8 model, the F1 confidence curve, why the F1 got dropped by 1.0 to 0.0 also it starts with 1.0? Why does it have to happen when there is an increase in the ...
0
votes
0
answers
12
views
AnswerRelevancyMetric not showing results on LLM evaluation
i've got this code :
from langchain_openai import AzureChatOpenAI
from deepeval.models.base_model import DeepEvalBaseLLM
class AzureOpenAI(DeepEvalBaseLLM):
def __init__(
self,
...
0
votes
0
answers
21
views
Micrometer metric max equals zero when sum is higher than zero
In our project we measure time for outgoing calls to other grpc servers, but I met a situation when a metric _max is zero, but the same metric for _sum is higher than zero.
@Slf4j
@...
0
votes
0
answers
14
views
How do I monitor metrics specific to Server Sent Events (SSE) in Spring?
We are using a SpringBoot service that uses SseEmitter to implement Server Sent Events back to the client.
From the standpoint of observability, I am wondering if there are any metrics that are ...
0
votes
0
answers
23
views
Setting up argocd metrics to get scraped at argocd-metrics:8082/metrics endpoint
Argocd v2.10 is managing applications but I need to update the templates to start scraping the metrics at the endpoint.
Application.metrics.enabled is already set to true.
How can the svc be created ...
-1
votes
1
answer
22
views
Proxmox metrics missing in exporter (PVE 8.2.4)
I installed the prometheus-pve-exporter on docker hosted by the proxmox. However, the metrics are missing. Only supporting items are available...
Such as:
pve_collection_duration_seconds_count{module=&...
0
votes
0
answers
11
views
Huggingface evaluate with f2
Is there a way to calculate f2 using huggingface's evaluate library?
ChatGPT instructed me to implement:
f2_metric = evaluate.load("f1", beta=2)
However, the output was identical to that of ...
0
votes
0
answers
21
views
event cli_task_run is 2 times more than count of event "running"
In this doc https://airflow.apache.org/docs/apache-airflow/stable/security/audit_logs.html I found
cli_task_run: Airflow triggered a task instance
I export metrics into Grafana throw this exporter ...
0
votes
0
answers
24
views
Discrepancy in RPS between custom Golang load generator and other tools
I'm working on a load generator written in Golang, and I'm currently using the Resty library to handle HTTP requests. The main logic of my load generator is fairly simple: I start goroutines that run ...
0
votes
0
answers
16
views
What timezone does the Google API use for segmenting metrics by date?
Consider this query for querying keyword metrics daily
https://developers.google.com/google-ads/api/docs/query/overview
SELECT
ad_group_criterion.criterion_id,
...
0
votes
0
answers
28
views
How to automatically obtain .net metrics in datadog using dogstatsD
We are using app.metrics to generate internal performance metrics for our applications' operations. These use the .Report.ToStatsDUdp to transmit to datadog's dogstatsD.
This works correctly.
We have ...
0
votes
0
answers
17
views
How to change the value of promethues internal label
I have an internal label on the target __meta_consul_service_port="8300"
I would like to set the value to 8501 in the relabel_configs
What I tried so far,
- job_name: 'conul'
...
0
votes
0
answers
17
views
Discrepancy between torch AUPRC and sklearn area under precision_recall_curve
I am writing a script that extends torch metrics to gives me some additional ease in using AUPRC and and AUROC in torch. Essentially I want them to be wrapped with a task parameter that allows me to ...
2
votes
0
answers
31
views
Prometheus Counter for time series aggregated data
I am using Prometheus Counter metrics, and each POST call, I increment the counter by custom amount. I am using Datadog to visualize the aggregated value by time interval, and I am having a trouble to ...
-1
votes
0
answers
19
views
MOT Occlusion Metric
Is there any Metric calculation such as IDF1, MOTA, HOTA, etc. specific just for calculate performance of the method / algorithm such as DeepSORT for occlusion object
I've tried IDF1, but is a metric ...