xiaohei/model_garden
1
0
Fork 0
mirror of https://github.com/GoogleCloudPlatform/vertex-ai-samples.git synced 2026-09-26 14:42:04 +00:00
Code Issues Packages Projects Releases Wiki Activity
2,654 Commits 1,765 Branches 0 Tags
a3bb4afe0677ec10322a7a2839aca213197764a3
Commit Graph
2654 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Mend RenovateandGitHub a3bb4afe06 chore(deps): update python docker tag to v3.12 (#2315) 2023-12-14 15:14:09 +00:00
Ryan LattureGitHubRyan Latture
 		f242e348c4 Document RLHF and inference pipelines support chat-bison@001. (#2577) 
This adds `chat-bison@001` to the table of supported models and documents chat dataset formats for relevant input parameters.

Co-authored-by: Ryan Latture <latture@google.com>
 2023-12-13 21:54:45 +00:00
weigaryandGitHub 8d47ab03c7 Add a new notebook for sdxl LCM (#2582) 
* Add SDXL-turbo as a separate section in the SDXL notebook.

* Add SDXL-turbo as a separate section in the SDXL notebook.

* Add a new notebook for sdxl LCM
 2023-12-13 21:50:35 +00:00
Minwoo ParkGitHubminwoopark
 		e8b86f503f Use open source detectron2 docker. (#2583) 
Co-authored-by: minwoopark <minwoopark@google.com>
 2023-12-13 21:48:35 +00:00
dstnluong-googleandGitHub 9449785b20 Switch TFvisio docker URIs to use vertex-ai (#2280) 2023-12-13 21:47:41 +00:00
genquan9andGitHub ac4f63b4e0 remove old vertex_vision_model_garden folder (#2570) 2023-12-13 19:56:38 +00:00
weigaryandGitHub 65ee5cdf52 Sd turbo (#2572) 
* Add SDXL-turbo as a separate section in the SDXL notebook.

* Add SDXL-turbo as a separate section in the SDXL notebook.
 2023-12-13 19:52:24 +00:00
dstnluong-googleandGitHub 90219e0a76 Add GPTQ to Falcon Instruct (#2574) 
* Add local inference for Falcon Instruct

* Lint

* Add memory necessary for falcon 7b

* nit: Typo

* add gptq quantization to falcon

* lint

* typo

* Comment out one setting.
 2023-12-13 19:49:39 +00:00
dstnluong-googleandGitHub d2ad1dd763 Add cache_examples_on_gpu flag to GPTQ. (#2579) 
* Add cache_examples_on_gpu flag to GPTQ.

* Lower GPU suggestions for 70b model.

* Lower GPU suggestions for 70b.

* Typo
 2023-12-13 19:47:07 +00:00
Kathy YuandGitHub b2ec0e956e Update timm notebook to use vertex-ai docker link. (#2580) 2023-12-13 19:45:01 +00:00
dstnluong-googleandGitHub 7fb60acf89 Create 5 separate notebooks from LLaMA2 notebook sections (#2568) 
* add deployment notebook

* lint

* add finetune notebook

* lint

* lint

* add evaluation notebook

* lint

* add rlhf notebook

* lint

* lint rlhf notebook?

* add notebooks to codeowners
 2023-12-05 20:50:46 +00:00
genquan9andGitHub 4f2408ce9e use the vertex_model_garden links instead of vertex_vision_model_garden (#2567) 2023-12-05 20:50:13 +00:00
Kathy YuandGitHub ef9a176c75 Add evaluation for CodeLLaMA and fix typos in CodeLLaMA and Mistral notebooks. (#2566) 2023-12-04 15:11:41 +00:00
genquan9andGitHub e72fffd682 copy vertex_vision_model_garden as vertex_model_garden (#2565) 2023-12-04 15:11:07 +00:00
KCFindstrandGitHub a9301eeb64 Add hyperparameter tuning examples to #PEFT Llama2 and OpenLlama notebooks (#2564) 2023-12-04 15:10:04 +00:00
weigaryandGitHub 9e2dcd494a Add charts to the benchmark report. (#2563) 
* Add charts to the benchmark report.

* Add charts to the benchmark report.
 2023-12-02 00:09:06 +00:00
KCFindstrandGitHub 2ae33b5b12 Add OpenLLaMA hyperparameter tuning benchmark report (#2562) 2023-12-01 22:47:03 +00:00
weigaryandGitHub d85a1e9c0c Update the SD XL notebook for refiner to use the default GPU as L4 (instead of A100). (#2561) 2023-12-01 22:46:30 +00:00
dstnluong-googleandGitHub fff9673703 Add SD benchmarking report (#2560) 
* Add SD benchmarking report

* Fix images

* remove old table

* grammar

* lint

* typo
 2023-12-01 22:45:50 +00:00
dstnluong-googleandGitHub 55fbb89a95 Update Docker URIs. (#2554) 
* Update Docker URIs.

* lint

* update docker URI

* typo

* add comma

* lint

* fix parameters

* Add quantization method to gptq
 2023-12-01 22:45:12 +00:00
dstnluong-googleandGitHub 007610f4f2 Add quantization section for Openllama (#2551) 
* Add quantization section for Openllama

* use peft serving for 3b quantized

* missing comma

* lint

* add quantization_method to gptq

* add quantization_method to gptq
 2023-12-01 22:44:36 +00:00
weigaryandGitHub 1354c74058 Move the openllama 7b benchmark report to the centralized place. (#2559) 2023-12-01 22:43:34 +00:00
Andrew FerlitschandGitHub f36ab8ef40 fix: online FS (#2558) 2023-11-30 17:08:43 +00:00
Andrew FerlitschandGitHub 33793dbb78 fix: backticks (#2557) 2023-11-30 15:05:18 +00:00
Lav RaiandGitHub d3bdf63115 Vit report (#2556) 
* Add ViT benchmarking report.

* Add ViT benchmarking report.

* Add ViT benchmarking report.

* Add ViT benchmarking report.

* Add ViT benchmarking report.

* Add ViT benchmarking report.
 2023-11-30 14:20:04 +00:00
weigaryandGitHub f3395041e4 Add openllama-7b fine tuning benchmark report. (#2555) 2023-11-30 14:19:29 +00:00
Andrew FerlitschandGitHub 5dfc87cbab fix: FS legacy delete (#2553) 2023-11-29 18:17:40 +00:00
Andrew FerlitschandGitHub a162066c94 in-progress: SDK whl replacement (#2552) 
* in-progress: SDK whl replacement

* add: cmd param
 2023-11-29 18:17:07 +00:00
dstnluong-googleandGitHub a4940cc2ba Add quantization to llama2 notebook (#2550) 2023-11-29 03:20:21 +00:00
Andrew FerlitschandGitHub d1dfe04ff0 2nd part of deprecate (#2542) 
* 2nd part of deprecate

* fix links
 2023-11-22 23:55:33 +00:00
Minwoo ParkGitHubminwoopark
 		fca8a0f9f5 Support DITO custom categories (#2543) 
Co-authored-by: minwoopark <minwoopark@google.com>
 2023-11-20 21:27:39 +00:00
Andrew FerlitschandGitHub ea3a5d3d79 deprecate flaky notebook (#2541) 2023-11-20 19:25:19 +00:00
Andrew FerlitschandGitHub 013bd9b78b fine tuning: auto error detection (#2539) 2023-11-20 18:42:28 +00:00
Andrew FerlitschandGitHub 1ca46e450d update: notebook status (#2535) 2023-11-20 16:52:16 +00:00
Xiang XuandGitHub 6c4f09fd43 increase max_num_batched_tokens for vllm and remove streaming (#2536) 2023-11-17 20:46:26 +00:00
Andrew FerlitschandGitHub 66427bd244 fix: detecting internal (#2534) 2023-11-17 20:37:13 +00:00
weigaryandGitHub d49bae8d29 Update staable-diffusion 1.5 and XL notebooks to deploy with default GPU as L4, instead of A100. (#2533) 2023-11-17 19:40:17 +00:00
Andrew FerlitschandGitHub 66c05c8217 fix: typo in results vs result (#2532) 2023-11-17 18:46:51 +00:00
Andrew FerlitschandGitHub 79473f0b88 fix: cleanup for FS 2.x (#2526) 2023-11-17 14:35:47 +00:00
Andrew FerlitschandGitHub 2568e012f9 fix: detect other types of quota and internal errors (#2525) 2023-11-16 23:50:25 +00:00
Kathy YuandGitHub 97fa323aca Fix typo in LLaMA 2 instructions. (#2524) 2023-11-16 22:04:52 +00:00
Kelsi LakeyandGitHub 20a9ff65e0 Update IMAGE_URI in tensorboard_custom_training_with_custom_container.ipynb (#2522) 
Add variable in IMAGE_URI to set region. Right now it is hard-coded to us-central1.
 2023-11-16 22:04:11 +00:00
Andrew FerlitschandGitHub ed6de9fde5 feat: stub and document reaper support for FS 2.x (#2521) 2023-11-15 22:04:01 +00:00
zhangxiaotianandGitHub dbaa3d098c fix clean up (#2520) 
* improve to make it more aligned with the template

* fix:fix clean up section
 2023-11-15 21:53:44 +00:00
Andrew FerlitschandGitHub 7e77de2800 CI/CD fix (#2514) 
* CI/CD fix

* pin 1.34
 2023-11-15 17:59:11 +00:00
Andrew FerlitschandGitHub ef8b70db32 Update CI/CD fix (#2513) 2023-11-15 00:09:33 +00:00
Kamal GuralaGitHubKamal Gurala
 		578adab3b0 revert how sdk is downloaded (#2518) 
Co-authored-by: Kamal Gurala <gkamal@google.com>
 2023-11-15 00:08:22 +00:00
KCFindstrandGitHub 3ef187ecc3 Update Llama2 PEFT notebook train docker version (#2519) 2023-11-15 00:05:45 +00:00
Kamal GuralaGitHubKamal Gurala
 		23dc370602 Updates optimized to remove known issues (#2512) 
Co-authored-by: Kamal Gurala <gkamal@google.com>
 2023-11-15 00:05:17 +00:00
Andrew FerlitschandGitHub d1835e98af feat: auto find latest build results (#2517) 2023-11-15 00:00:31 +00:00