-
Notifications
You must be signed in to change notification settings - Fork 33
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Base models and notebooks for general IPU embeddings model #436
Merged
katalinic-gc
merged 31 commits into
huggingface:main
from
graphcore:text_embeddings_models
Jul 14, 2023
Merged
Base models and notebooks for general IPU embeddings model #436
katalinic-gc
merged 31 commits into
huggingface:main
from
graphcore:text_embeddings_models
Jul 14, 2023
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. |
3 tasks
…dd required model classes to source
… incompatible ipu configs
relies on #437 being merged first |
…-graphcore-fork into text_embeddings_models
jimypbr
reviewed
Jul 6, 2023
…/optimum-graphcore-fork into text_embeddings_models
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Support for multiple embeddings models (e5, mpnet, t5 (pending))
Idea is to create a general embeddings class which supports the pool and norm stage of embeddings model. These models use encoder stages from various base models (bert, t5, mpnet) which we don't support the direct use of with pipelining. I've added these in this folder for now, but they may need to be in the source lib if integrated.
Seemed like not a great idea to create new pipelined classes for all of these models, as they inherit from existing models we support and would just be an extra couple of lines which are common amongst them, so makes more sense to have a generic wrapper for embeddings, which instantiates the encoder stages through AutoModel as currently shown in HF.
The inference setup is manually outlined, but may be more user-friendly to create a pipeline for these eventually.
MPNet modeling file only includes the base model and maskedLM model, not tested for pretraining, only for embeddings.
notebook will likely be restructured