lplcor

Reason for still using "default" in revision name?

4

In kfserving v0.4 we have primary route with name `{isvc_name}-predictor-default` and canary route with name `{isvc_name}-predictor-canary`, which makes sense to distinguish them. However in v0.5, from what I see [here](https://github.com/kubeflow/kfserving/tree/master/docs/samples/v1beta1/torchserve/canary#check-the-traffic-split-between-the-two-revisions),...

/kind feature **Describe the solution you'd like** [A clear and concise description of what you want to happen.] First thanks for the work on [tag based routing](https://github.com/kserve/kserve/tree/master/docs/samples/v1beta1/rollout#tag-based-routing). We are exploring...

Improve Python SDK dependency management strategy

1

/kind feature **Describe the solution you'd like** [A clear and concise description of what you want to happen.] As discussed in WG meeting ([notes](https://docs.google.com/document/d/1KZUURwr9MnHXqHA08TFbfVbM8EAJSJjmaMhnvstvi-k/edit#heading=h.x518nkl3k9ww)), it'd be better if we could...

/kind feature **Describe the solution you'd like** [A clear and concise description of what you want to happen.] This is a minor request, compared to the main one https://github.com/kserve/kserve/issues/2403 Compared...

### Which version of dd-trace-py are you using? 0.49.4 ### Which version of the libraries are you using? tornado = 6.1 ### How can we reproduce your problem? We are...

Autoscaler underprovisions for uneven low latency traffic

3

## Ask your question here: Hi community, we have a potentially skewed low latency traffic targeting a CPU-bound knative service. With concurrency-based autoscaling, we are seeing a high p90+ latency....

Activator will OOM under backpressure

8

## What version of Knative? 0.22.3 ## Expected Behavior Activator should be resilient under backpressure ## Actual Behavior Activator OOM'ed under backpressure ## Steps to Reproduce the Problem 1. Create...

Thanks for the project! We want to run lightllm directly in a cloud container environment where the current way to provide a `model_dir` is harder than providing a huggingface model...

Consideration behind a "stateful" metric UX

9

**Is your feature request related to a problem? Please describe.** Hi, I was wondering what's the considerations behind that we are choosing a "stateful" metrics UX. Using the hallucination metric...

Summarization metric generates <1 score for exactly the same input and output

1

**Describe the bug** (First thanks for the awesome work! We found a potential analytical caveat for the summarization metric during our prototype and want to report it in case it...