Eero Tamminen
Eero Tamminen
**What are you trying to achieve?** Follow recommendations on: https://opentelemetry.io/docs/specs/semconv/system/hardware-metrics/ **What did you expect to see?** Consistent and implementable specification. **Additional context.** Ran across following issues when trying to map...
Building latest `compute-runtime` tag (`24.22.29735.15`) fails: ``` $ cmake -LH -Wno-dev -G Ninja -DCMAKE_INSTALL_PREFIX=/usr/local -DCMAKE_BUILD_TYPE=Release -DSUPPORT_GEN8=0 -DSUPPORT_GEN9=1 -DSUPPORT_GEN11=0 -DSUPPORT_TGLLP=1 -DSUPPORT_DG1=1 -DSUPPORT_XE_HP_SDV=1-DSUPPORT_DG2=1 -DSUPPORT_PVC=1 -DNEO_ENABLE_i915_PRELIM_DETECTION=TRUE -DNEO_DISABLE_LD_GOLD=1 -DDO_NOT_RUN_AUB_TESTS=1 -DDONT_CARE_OF_VIRTUALS=1 $ ninja ... #21...
The model used for ChatQnA supports BFLOAT16, in addition to TGI's default 32-bit float type: https://huggingface.co/Intel/neural-chat-7b-v3-3 TGI memory usage halves from 30GB to 15GB (and also its perf increases somewhat)...
Many of the `Dockerfile`s install Vim and/or Mesa OpenGL/X packages: ``` $ git grep -l -B1 -e mesa-glx -e '\bvim\b' AudioQnA/langchain/docker/Dockerfile ChatQnA/deprecated/langchain/docker/Dockerfile ChatQnA/docker/Dockerfile CodeGen/deprecated/codegen/Dockerfile CodeGen/docker/Dockerfile CodeTrans/deprecated/langchain/docker/Dockerfile CodeTrans/docker/Dockerfile DocSum/deprecated/langchain/docker/Dockerfile DocSum/docker/Dockerfile Translation/langchain/docker/Dockerfile...
Many docs in this repo instruct giving HTTP/S proxies on Docker build command line: ``` $ git grep -e "--build-arg.*https*_proxy=" | wc -l 58 ``` IMHO it would be better...
**Setup** These errors originally happened with v0.7 ChatQnA Xeon installation [1], but e.g. updating to TEI services from `1.2-cpu` version to latest `1.5-cpu`, and and TGI service from `1.4` version...
Currently one can get inferencing metrics from TGI and TEI backend services, but there are no E2E metrics for the whole pipeline, e.g. what are the first response, and response...
I would expect seeing pod container `securityContext`s like this: ``` securityContext: allowPrivilegeEscalation: false readOnlyRootFilesystem: true seccompProfile: type: RuntimeDefault capabilities: drop: [ "ALL" ] ``` And `runAsUser` setting for something else...
Continuation of #361. Currently OPEA images in DockerHub are lacking descriptions: https://hub.docker.com/u/opea Would be good to describe the purpose of the image, components included to it and their licensing (BOM),...
OPEA services fetch "random" data from internet (HF), and try to access each others' k8s service end points. Access to Internet from an intranet can be handled by adding `http_proxy`/`https_proxy`...