Hugging Face's logo
Aditya Gupta's picture
1 4

Aditya Gupta

Adi-0-0-Gupta
https://www.linkedin.com/in/adi-iitd/

AI & ML interests

Machine Learning, Deep Learning, NLP, Computer Vision, Generative models

Recent Activity

reacted to tomaarsen's post with šŸ˜Ž 3 days ago
I just released Sentence Transformers v3.4.0, featuring a memory leak fix, compatibility between the powerful Cached... losses and the Matryoshka loss modifier, and a bunch of fixes & small features. šŸŖ† Matryoshka & Cached loss compatibility It is now possible to combine the powerful Cached... losses (which use in-batch negatives & a caching mechanism to allow for endless batch size & negatives) with the Matryoshka loss modifier which modifies a base loss such that it is trained not only on the maximum dimensionality (e.g. 1024 dimensions), but also on many lower dimensions (e.g. 768, 512, 256, 128, 64, 32). After training, these models' embeddings can be truncated for faster retrieval, etc. šŸŽžļø Resolve memory leak when Model and Trainer are reinitialized Due to a circular dependency between Trainer -> Model -> ModelCardData -> Trainer, deleting both the trainer & model still didn't free up the memory. This led to a memory leak in scripts where you repeatedly do so. āž• New Features Many new small features, e.g. multi-GPU support for 'mine_hard_negatives', a 'margin' parameter to TripletEvaluator, and Matthews Correlation Coefficient in the BinaryClassificationEvaluator. šŸ› Bug Fixes Also a bunch of fixes, for example that subsequent batches were not sorted when using the "no_duplicates" batch sampler. See the release notes for more details. Full release notes: https://github.com/UKPLab/sentence-transformers/releases/tag/v3.4.0 Big thanks to all community members who assisted in this release. 10 folks with their first contribution this time around!
updated a model about 1 month ago
Adi-0-0-Gupta/llama-3b-recipe-simplification
upvoted an article 7 months ago
Welcome Gemma 2 - Google's new open LLM
View all activity

Organizations

None yet