jandres/GenAIExamples
1
0
Fork 0
Code Issues Pull Requests Actions 5 Packages Projects Releases Wiki Activity
870 Commits 43 Branches 9 Tags
c8c6fa2e3eb243bdfa70321ab5b10062c529d3ff
Commit Graph

870 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
bjzhjing
c8c6fa2e3e Provide unified scalable deployment and benchmarking support for exam… (#1315) 
Signed-off-by: Cathy Zhang <cathy.zhang@intel.com>
Signed-off-by: letonghan <letong.han@intel.com>
Co-authored-by: letonghan <letong.han@intel.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
(cherry picked from commit ed163087ba)
v1.2 		2025-01-24 22:55:38 +08:00
NeuralChatBot
905a5100f9 Freeze OPEA images tag 
Signed-off-by: NeuralChatBot <grp_neural_chat_bot@intel.com>
 2025-01-24 08:31:22 +00:00
chen, suyue
259099d19f Remove kubernetes manifest related code and tests (#1466) 
Remove deprecated kubernetes manifest related code and tests.
k8s implementation for those examples based on helm charts will target for next release.

Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-24 15:23:12 +08:00
chen, suyue
9a1118730b Freeze the triton version in vllm-gaudi image to 3.1.0 (#1463) 
The new triton version 3.2.0 can't work with vllm-gaudi. Freeze the triton version in vllm-gaudi image to 3.1.0.

Issue create for vllm-fork: HabanaAI/vllm-fork#732
Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-24 09:50:59 +08:00
chen, suyue
ffce7068aa Fix image on push action due to manifest test remove (#1460) 
1. Fix image on push action due to manifest test remove.
2. Fix helm test cd workflow get test matrix step.
Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-23 14:30:09 +08:00
dolpher
9b0f98be8b Update ChatQnA helm chart README. (#1459) 
Signed-off-by: Dolpher Du <dolpher.du@intel.com>
 2025-01-23 10:54:39 +08:00
XinyuYe-Intel
f0fea7b706 Add docker compose yaml for text2image example (#1418) 
Signed-off-by: Ye, Xinyu <xinyu.ye@intel.com>
 2025-01-23 09:57:54 +08:00
Melanie Hart Buehler
1864fac978 Fixes MultimodalQnA dataprep endpoint and port in the UI (#1457) 
Signed-off-by: Melanie Buehler <melanie.h.buehler@intel.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
 2025-01-22 17:11:09 -08:00
Lianhao Lu
94f71f2322 Update top level readme (#1458) 
Add helm support of SeachQnA and Text2Image in top level readme.

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
 2025-01-23 09:07:33 +08:00
chen, suyue
6600c32a9b remove image build condition (#1456) 
Test compose cd workflow depend on image build, so if we want to run both compose and helm charts deployment in cd workflow, this condition should be removed.

Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-23 00:17:04 +08:00
Liang Lv
d953332f43 Fix multimodal docker image issue for MutimodalQnA on Gaudi (#1455) 
Signed-off-by: lvliang-intel <liang1.lv@intel.com>
 2025-01-23 00:12:06 +08:00
chyundunovDatamonsters
cbe5805f47 AgentQnA - add README file for deploy on ROCm (#1379) 
Signed-off-by: Chingis Yundunov <YundunovCN@sibedge.com>
 2025-01-22 21:57:15 +08:00
Ervin Castelino
27fdbcab58 [chore/chatqna] Missing protocol in curl command (#1447) 
This PR fixes the missing protocol for in the curl command mentioned in chatqna readme for tei-embedding-service.
 2025-01-22 21:41:47 +08:00
lkk
f07cf1dad2 Fix wrong vllm repo. (#1454) 
Use vllm-fork for gaudi.

fix the issue #1451
 2025-01-22 21:22:56 +08:00
dolpher
ee0e5cc8d9 Sync value files from GenAIInfra (#1428) 
All gaudi values updated with extra flags.
Added helm support for 2 new examples Text2Image and SearchQnA. Minor fix for llm-uservice.

Signed-off-by: Dolpher Du <dolpher.du@intel.com>
 2025-01-22 17:44:11 +08:00
chen, suyue
5c36443b11 Use local hub cache for AgentQnA test (#1450) 
Use local hub cache for AgentQnA test to save workspace.

Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-22 13:23:00 +08:00
Lianhao Lu
62cea74a23 CI: improve helm CI (#1452) 
Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
 2025-01-22 09:18:35 +08:00
WenjiaoYue
b721c256f9 Fix Domain Access Issue in Latest Vite Version (#1444) 
Fix the restriction on using domain names when users are using the latest version of Vite

When users use the new version of Vite, the UI cannot be accessed via domain names due to Vite's new rules. This fix adds the corresponding parameters according to Vite's new rules, ensuring that users can access the frontend via domain names when building the UI.

Fixes #1441

Co-authored-by: WenjiaoYue <wenjiao.yue@intel.com>
 2025-01-21 23:28:37 +08:00
chen, suyue
927698e23e Simplify git clone code in CI test (#1434) 
1. Simplify git clone code in CI test.
2. Replace git clone branch in Dockerfile.

Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-21 23:00:08 +08:00
ZePan110
c3e84b5ffa Fix test matrix for helm charts (#1449) 
Signed-off-by: ZePan110 <ze.pan@intel.com>
 2025-01-21 22:28:31 +08:00
ZePan110
6b2a041f25 Fix Helm-chart workflow issues. (#1448) 
Fix matrix error issues and CD test files cannot be obtained.

Signed-off-by: ZePan110 <ze.pan@intel.com>
 2025-01-21 21:48:57 +08:00
ZePan110
842f46326b Switch helm-chart test runs-on label. (#1446) 
Switch helm-chart test runs-on label from ${{ inputs.hardware }} to k8s-${{ inputs.hardware }}.

Signed-off-by: ZePan110 <ze.pan@intel.com>
 2025-01-21 18:07:03 +08:00
Wang, Kai Lawrence
284db982be [ROCm] Fix the hf-token setting for TGI and TEI in ChatQnA (#1432) 
This PR is to correct the env variable names in chatqna example on ROCm platform passing to the docker container of TGI and TEI. For tgi, either HF_TOKEN and HUGGING_FACE_HUB_TOKEN could be parsed in TGI while HF_API_TOKEN can be parsed in TEI.

TGI: https://github.com/huggingface/text-generation-inference/blob/main/router/src/server.rs#L1700C1-L1702C15
TEI: https://github.com/huggingface/text-embeddings-inference/blob/main/router/src/main.rs#L112

Signed-off-by: Wang, Kai Lawrence <kai.lawrence.wang@intel.com>
 2025-01-21 14:22:39 +08:00
ZePan110
fc96fe83e2 Fix CD workflow issue (#1443) 
Fix the issue of CD workflow values_files errors.

Signed-off-by: ZePan110 <ze.pan@intel.com>
 2025-01-21 11:54:12 +08:00
Hoong Tee, Yeoh
0316114c4b ProductivitySuite: Fix FaqGen Microservice CI test fail (#1437) 
Change in FAQGen microservice for content-type header result in CI failure.

#1431
Signed-off-by: Yeoh, Hoong Tee <hoong.tee.yeoh@intel.com>
 2025-01-21 10:23:35 +08:00
chen, suyue
0408453fa2 Unify the yaml name to fix the CD workflow (#1435) 
Fix the issue in #1372

Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-21 01:10:41 +08:00
XinyaoWa
d0cd0aaf53 Update GraphRAG to be compatible with latest component changes (#1427) 
- Updated ENV VARS to align with recent changes in neo4j dataprep and retriever.
- upgraded tgi-gaudi image version
Related to GenAIComps repo issue #1025 (opea-project/GenAIComps#1025)

Original PR #1384
Original contributor is @rbrugaro

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
Co-authored-by: Liang Lv <liang1.lv@intel.com>
 2025-01-21 00:18:01 +08:00
chen, suyue
0ba3decb6b Simplify git clone code in CI test (#1422) 
1. Simplify git clone code in CI test. 
2. Replace git clone branch in Dockerfile.

Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-20 23:55:20 +08:00
Wang, Kai Lawrence
3d3ac59bfb [ChatQnA] Update the default LLM to llama3-8B on cpu/gpu/hpu (#1430) 
Update the default LLM to llama3-8B on cpu/nvgpu/amdgpu/gaudi for docker-compose deployment to avoid the potential model serving issue or the missing chat-template issue using neural-chat-7b.

Slow serving issue of neural-chat-7b on ICX: #1420
Signed-off-by: Wang, Kai Lawrence <kai.lawrence.wang@intel.com>
 2025-01-20 22:47:56 +08:00
Melanie Hart Buehler
f11ab458d8 MultimodalQnA image query, pdf, dynamic ports, and UI updates (#1381) 
Per the proposed changes in this [RFC](https://github.com/opea-project/docs/blob/main/community/rfcs/24-10-02-GenAIExamples-001-Image_and_Audio_Support_in_MultimodalQnA.md)'s Phase 2 plan, this PR adds support for image queries, PDF ingestion and display, and dynamic ports. There are also some bug fixes. This PR goes with [this one in GenAIComps](https://github.com/opea-project/GenAIComps/pull/1134).

Signed-off-by: Melanie Buehler <melanie.h.buehler@intel.com>
Co-authored-by: Liang Lv <liang1.lv@intel.com>
 2025-01-20 22:41:52 +08:00
ZePan110
f3562bef36 Add helm e2e test workflow (#1372) 
Add both CICD workflow for helm charts values test. 

Signed-off-by: ZePan110 <ze.pan@intel.com>
 2025-01-20 21:04:11 +08:00
chen, suyue
7a54064d65 remove Dockerfile.wrapper (#1429) 
Remove Dockerfile.wrapper, it's not used anymore and no test cover this Dockerfile. So remove this Dockerfile to avoid regression.

Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-20 20:49:18 +08:00
Liang Lv
0f7e5a37ac Adapt code for dataprep microservice refactor (#1408) 
https://github.com/opea-project/GenAIComps/pull/1153

Signed-off-by: lvliang-intel <liang1.lv@intel.com>
 2025-01-20 20:37:03 +08:00
xiguiw
2d5898244c Enchance health check in GenAIExample docker-compose (#1410) 
Fix service launch issue

1. Update Gaudi TGI image from 2.0.6 to 2.3.1
2. Change the hpu-gaudi TGI health check condition.

Signed-off-by: Wang, Xigui <xigui.wang@intel.com>
 2025-01-20 20:13:13 +08:00
Neo Zhang Jianyu
59722d2bc9 [Bug] Enhance the template (#1396) 
Enhance the bug & feature template according to the issue #1002.
Co-authored-by: ZhangJianyu <zhang.jianyu@outlook.com>
 2025-01-20 17:56:14 +08:00
chen, suyue
6bfd156573 Clean up test scripts and enhance git clone (#1417) 
1. Clean up test code in scripts.
2. Simplify git clone code.
3. Replace git clone branch in Dockerfile.

Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-20 16:34:28 +08:00
XinyuYe-Intel
528770a8d7 Add UT for Text2Image on Gaudi (#1424) 
Add UT for Text2Image on Gaudi.

#1421
Signed-off-by: Ye, Xinyu <xinyu.ye@intel.com>
 2025-01-20 16:01:35 +08:00
chen, suyue
239995da16 Update DocIndexRetriever CI test scripts (#1416) 
1. Add image build condition.
2. Update single branch clone.

Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-20 11:16:38 +08:00
chen, suyue
f65e8d8668 Add port 5000 checking and warning (#1414) 
Port 5000 is used by local docker registry, please DO NOT use it in docker compose deployment!!!

Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-20 09:09:31 +08:00
chen, suyue
a49a36cebc Add secrets OPENAI_API_KEY (#1412) 
Add secrets OPENAI_API_KEY for AMD GPU CI test. 

Signed-off-by: chensuyue <suyue.chen@intel.com>
 2025-01-19 19:39:45 +08:00
Wang, Kai Lawrence
742cb6ddd3 [ChatQnA] Switch to vLLM as default llm backend on Xeon (#1403) 
Switching from TGI to vLLM as the default LLM serving backend on Xeon for the ChatQnA example to enhance the perf.

https://github.com/opea-project/GenAIExamples/issues/1213
Signed-off-by: Wang, Kai Lawrence <kai.lawrence.wang@intel.com>
 2025-01-17 20:48:19 +08:00
Wang, Kai Lawrence
00e9da9ced [ChatQnA] Switch to vLLM as default llm backend on Gaudi (#1404) 
Switching from TGI to vLLM as the default LLM serving backend on Gaudi for the ChatQnA example to enhance the perf. 

https://github.com/opea-project/GenAIExamples/issues/1213
Signed-off-by: Wang, Kai Lawrence <kai.lawrence.wang@intel.com>
 2025-01-17 20:46:38 +08:00
chyundunovDatamonsters
277222a922 General README.md - add deploy on AMD info (#1409) 
Signed-off-by: Chingis Yundunov <YundunovCN@sibedge.com>
Co-authored-by: Chingis Yundunov <YundunovCN@sibedge.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
 2025-01-17 20:26:59 +08:00
lkk
5c68effc9f update agent example for the GenAIComps changes. (#1407) 
Update build.yaml and compose_vllm.yaml because of refactoring of GenAIComps.

Fix issue left by https://github.com/opea-project/GenAIExamples/pull/1353
 2025-01-17 11:29:11 +08:00
XinyaoWa
39409d7f61 Align OpenAI API for FaqGen, DocSum (#1401) 
Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
 2025-01-17 11:19:35 +08:00
XinyaoWa
71e3c57366 Standardize name for LLM comps (#1402) 
Update all the names for classes and files in llm comps to follow the standard format, related GenAIComps PR opea-project/GenAIComps#1162

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
 2025-01-16 23:10:27 +08:00
Letong Han
5ad24af2ee Fix Vectorestores Path Issue of Refactor (#1399) 
Fix vectorestores path issue caused by refactor in PR opea-project/GenAIComps#1159.
Modify docker image name and file path in docker_images_list.md.

Signed-off-by: letonghan <letong.han@intel.com>
 2025-01-16 19:50:59 +08:00
WenjiaoYue
3a9a24a51a Agent ui (#1389) 
Signed-off-by: WenjiaoYue <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>
Co-authored-by: WenjiaoYue <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>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
 2025-01-16 18:47:46 +08:00
XinyaoWa
301b5e9a69 Fix vllm hpu to a stable release (#1398) 
Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
 2025-01-16 16:35:32 +08:00
Yao Qing
b4269d6c4f Modify the corresponding path based on the refactor of chathistory in GenAIComps. (#1397) 
GenAIComps has refactored chathistory based on E-RAG code structure. Related path in GenAIExample have been modified.

Fix GenAIComps Issue https://github.com/opea-project/GenAIComps/issues/989 
Signed-off-by: Yao, Qing <qing.yao@intel.com>
 2025-01-16 14:26:17 +08:00